993 lines
88 KiB
Plaintext
993 lines
88 KiB
Plaintext
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 1,
|
|
"id": "7174c11a",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"import polars as pl\n",
|
|
"from utils import QualtricsSurvey, calculate_weighted_ranking_scores"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 2,
|
|
"id": "d9d11d52",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"SPEECH_DATA = 'data/speech_data/JPMC Speech data ab samples (Final Speech Data AB samples).csv'\n",
|
|
"RESULTS_FILE = 'data/exports/2-4-26/JPMC_Chase Brand Personality_Quant Round 1_February 4, 2026_Labels.csv'\n",
|
|
"QSF_FILE = 'data/exports/OneDrive_2026-01-21/Soft Launch Data/JPMC_Chase_Brand_Personality_Quant_Round_1.qsf'"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 37,
|
|
"id": "c8f06ff8",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"speech_df = pl.read_csv(SPEECH_DATA, separator=';')\n",
|
|
"\n"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"id": "8da85898",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": []
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 38,
|
|
"id": "e7ccd8ef",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"\n",
|
|
"# Convert comma decimal separator to period and cast to float\n",
|
|
"cols_to_convert = [\n",
|
|
" pl.col('dur (s)').str.replace(',', '.').cast(pl.Float64),\n",
|
|
" pl.col('phonationtime (s)').str.replace(',', '.').cast(pl.Float64),\n",
|
|
" pl.col('articulation rate (nsyll / phonationtime)').str.replace(',', '.').cast(pl.Float64),\n",
|
|
" pl.col('speech rate words per minute').str.replace(',', '.').cast(pl.Float64),\n",
|
|
"]\n",
|
|
"\n",
|
|
"# Convert specified columns to float, handling percent signs, commas and spaces\n",
|
|
"if \"Standard deviation pitch\" in speech_df.columns:\n",
|
|
" cols_to_convert.append(\n",
|
|
" pl.col(\"Standard deviation pitch\")\n",
|
|
" .cast(pl.Utf8)\n",
|
|
" .str.replace(\",\", \".\")\n",
|
|
" .str.replace(\" \", \"\")\n",
|
|
" # strict=False converts unparseable strings (like empty ones) to Null\n",
|
|
" .cast(pl.Float64, strict=False)\n",
|
|
" )\n",
|
|
"\n",
|
|
"for col_name in [\"Jitter (local)\", \"Shimmer (local)\"]:\n",
|
|
" if col_name in speech_df.columns:\n",
|
|
" cols_to_convert.append(\n",
|
|
" pl.col(col_name)\n",
|
|
" .cast(pl.Utf8)\n",
|
|
" .str.replace(\"%\", \"\")\n",
|
|
" .str.replace(\" \", \"\")\n",
|
|
" .str.replace(\",\", \".\")\n",
|
|
" .cast(pl.Float64, strict=False)\n",
|
|
" )\n",
|
|
"\n",
|
|
"if cols_to_convert:\n",
|
|
" speech_df = speech_df.with_columns(cols_to_convert)\n",
|
|
"\n",
|
|
"# speech_df\n"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 39,
|
|
"id": "450d1d29",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"shape: (18, 3)\n",
|
|
"┌───────┬────────┬──────┐\n",
|
|
"│ Voice ┆ Gender ┆ Age │\n",
|
|
"│ --- ┆ --- ┆ --- │\n",
|
|
"│ str ┆ str ┆ i64 │\n",
|
|
"╞═══════╪════════╪══════╡\n",
|
|
"│ V04 ┆ female ┆ 28 │\n",
|
|
"│ V08 ┆ female ┆ 45 │\n",
|
|
"│ V14 ┆ female ┆ 50 │\n",
|
|
"│ V16 ┆ male ┆ 40 │\n",
|
|
"│ V34 ┆ male ┆ 42 │\n",
|
|
"│ … ┆ … ┆ … │\n",
|
|
"│ V82 ┆ female ┆ null │\n",
|
|
"│ V86 ┆ male ┆ 62 │\n",
|
|
"│ V88 ┆ male ┆ 42 │\n",
|
|
"│ V89 ┆ female ┆ 32 │\n",
|
|
"│ V91 ┆ female ┆ null │\n",
|
|
"└───────┴────────┴──────┘\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"# Convert Voice ints to zero-padded strings like \"V04\", \"V81\"\n",
|
|
"# Survey uses zero-padded IDs (V04, V08) so we must match that format\n",
|
|
"speech_df = speech_df.with_columns(\n",
|
|
" pl.when(pl.col(\"Voice\").is_not_null())\n",
|
|
" .then(pl.concat_str([pl.lit(\"V\"), pl.col(\"Voice\").cast(pl.Utf8).str.zfill(2)]))\n",
|
|
" .otherwise(None)\n",
|
|
" .alias(\"Voice\")\n",
|
|
")\n",
|
|
"\n",
|
|
"print(speech_df.select([\"Voice\", \"Gender\", \"Age\"]).sort(\"Voice\"))"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "5fb615fe",
|
|
"metadata": {},
|
|
"source": [
|
|
"# Get survey data"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 7,
|
|
"id": "bb4200ee",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"S = QualtricsSurvey(RESULTS_FILE, QSF_FILE, figures_dir=None)\n",
|
|
"data_all = S.load_data()\n",
|
|
"data = S.filter_data(data_all)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 8,
|
|
"id": "57243afd",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"shape: (5, 2)\n",
|
|
"┌───────────┬────────────────┐\n",
|
|
"│ Character ┆ Weighted Score │\n",
|
|
"│ --- ┆ --- │\n",
|
|
"│ str ┆ i64 │\n",
|
|
"╞═══════════╪════════════════╡\n",
|
|
"│ V14 ┆ 209 │\n",
|
|
"│ V04 ┆ 209 │\n",
|
|
"│ V08 ┆ 180 │\n",
|
|
"│ V82 ┆ 172 │\n",
|
|
"│ V77 ┆ 158 │\n",
|
|
"└───────────┴────────────────┘\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"top3_voices = S.get_top_3_voices(data)[0]\n",
|
|
"voices_weighted_rank = calculate_weighted_ranking_scores(top3_voices)\n",
|
|
"print(voices_weighted_rank.head())"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 9,
|
|
"id": "b38d21fc",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"shape: (5, 18)\n",
|
|
"┌───────────┬───────────┬───────────┬───────────┬───┬───────────┬───────────┬───────────┬──────────┐\n",
|
|
"│ _recordId ┆ Voice_Sca ┆ Voice_Sca ┆ Voice_Sca ┆ … ┆ Voice_Sca ┆ Voice_Sca ┆ Voice_Sca ┆ Voice_Sc │\n",
|
|
"│ --- ┆ le_1_10__ ┆ le_1_10__ ┆ le_1_10__ ┆ ┆ le_1_10__ ┆ le_1_10__ ┆ le_1_10__ ┆ ale_1_10 │\n",
|
|
"│ str ┆ V14 ┆ V04 ┆ V08 ┆ ┆ V74 ┆ V81 ┆ V86 ┆ __V88 │\n",
|
|
"│ ┆ --- ┆ --- ┆ --- ┆ ┆ --- ┆ --- ┆ --- ┆ --- │\n",
|
|
"│ ┆ f64 ┆ f64 ┆ f64 ┆ ┆ f64 ┆ f64 ┆ f64 ┆ f64 │\n",
|
|
"╞═══════════╪═══════════╪═══════════╪═══════════╪═══╪═══════════╪═══════════╪═══════════╪══════════╡\n",
|
|
"│ R_59pdrC3 ┆ null ┆ null ┆ null ┆ … ┆ null ┆ null ┆ 5.5 ┆ null │\n",
|
|
"│ urLmZnbP ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │\n",
|
|
"│ R_3fJSKy5 ┆ 6.0 ┆ 5.0 ┆ null ┆ … ┆ null ┆ null ┆ 6.0 ┆ null │\n",
|
|
"│ SVxmNdBC ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │\n",
|
|
"│ R_3g11G0u ┆ 9.5 ┆ null ┆ 5.0 ┆ … ┆ null ┆ null ┆ null ┆ 9.5 │\n",
|
|
"│ pJ7iGt8Q ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │\n",
|
|
"│ R_3i3dGL7 ┆ 6.0 ┆ 9.0 ┆ 8.0 ┆ … ┆ null ┆ 2.0 ┆ 3.0 ┆ null │\n",
|
|
"│ cfLOTgxb ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │\n",
|
|
"│ R_3BBF1fR ┆ null ┆ null ┆ null ┆ … ┆ 6.0 ┆ null ┆ 8.5 ┆ null │\n",
|
|
"│ WGGeButr ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │\n",
|
|
"└───────────┴───────────┴───────────┴───────────┴───┴───────────┴───────────┴───────────┴──────────┘\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"voice_1_10 = S.get_voice_scale_1_10(data)[0].collect()\n",
|
|
"print(voice_1_10.head())\n",
|
|
"\n"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 10,
|
|
"id": "5b3e6ad0",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"shape: (5, 2)\n",
|
|
"┌───────┬────────────────────────┐\n",
|
|
"│ Voice ┆ Avg Voice Score (1-10) │\n",
|
|
"│ --- ┆ --- │\n",
|
|
"│ str ┆ f64 │\n",
|
|
"╞═══════╪════════════════════════╡\n",
|
|
"│ V08 ┆ 7.38172 │\n",
|
|
"│ V82 ┆ 7.376984 │\n",
|
|
"│ V89 ┆ 7.373206 │\n",
|
|
"│ V86 ┆ 7.264444 │\n",
|
|
"│ V69 ┆ 7.219577 │\n",
|
|
"└───────┴────────────────────────┘\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"# --- Compute average voice score (1-10) per voice ---\n",
|
|
"voice_cols = [c for c in voice_1_10.columns if c.startswith(\"Voice_Scale_1_10__\")]\n",
|
|
"avg_scores = []\n",
|
|
"for col in voice_cols:\n",
|
|
" voice_id = col.replace(\"Voice_Scale_1_10__\", \"\") # e.g. \"V14\"\n",
|
|
" mean_val = voice_1_10.select(pl.col(col).mean()).item()\n",
|
|
" avg_scores.append({\"Voice\": voice_id, \"Avg Voice Score (1-10)\": mean_val})\n",
|
|
"\n",
|
|
"avg_voice_scores = pl.DataFrame(avg_scores)\n",
|
|
"print(avg_voice_scores.sort(\"Avg Voice Score (1-10)\", descending=True).head())"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 40,
|
|
"id": "79626ffb",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Merged rows: 18 (voices with both speech data and survey data)\n",
|
|
" → Voices missing Avg Voice Score: ['V46']\n",
|
|
"shape: (5, 4)\n",
|
|
"┌───────┬────────┬────────────────────────┬────────────────┐\n",
|
|
"│ Voice ┆ Gender ┆ Avg Voice Score (1-10) ┆ Weighted Score │\n",
|
|
"│ --- ┆ --- ┆ --- ┆ --- │\n",
|
|
"│ str ┆ str ┆ f64 ┆ i64 │\n",
|
|
"╞═══════╪════════╪════════════════════════╪════════════════╡\n",
|
|
"│ V14 ┆ female ┆ 7.216279 ┆ 209 │\n",
|
|
"│ V04 ┆ female ┆ 7.07971 ┆ 209 │\n",
|
|
"│ V08 ┆ female ┆ 7.38172 ┆ 180 │\n",
|
|
"│ V82 ┆ female ┆ 7.376984 ┆ 172 │\n",
|
|
"│ V77 ┆ female ┆ 6.960894 ┆ 158 │\n",
|
|
"└───────┴────────┴────────────────────────┴────────────────┘\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"# --- Normalize weighted rank column name and join all data ---\n",
|
|
"weighted_rank = voices_weighted_rank.rename({\"Character\": \"Voice\"})\n",
|
|
"\n",
|
|
"# Join speech attributes with both survey metrics\n",
|
|
"# Left join on avg_voice_scores so V46 (excluded from survey voice scale) is kept\n",
|
|
"# — its Avg Score will be null but Weighted Ranking Score is still valid\n",
|
|
"merged = (\n",
|
|
" speech_df\n",
|
|
" .join(avg_voice_scores, on=\"Voice\", how=\"left\")\n",
|
|
" .join(weighted_rank, on=\"Voice\", how=\"inner\")\n",
|
|
")\n",
|
|
"\n",
|
|
"print(f\"Merged rows: {merged.height} (voices with both speech data and survey data)\")\n",
|
|
"print(f\" → Voices missing Avg Voice Score: {merged.filter(pl.col('Avg Voice Score (1-10)').is_null())['Voice'].to_list()}\")\n",
|
|
"print(merged.select([\"Voice\", \"Gender\", \"Avg Voice Score (1-10)\", \"Weighted Score\"]).head())"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "932cfb9e",
|
|
"metadata": {},
|
|
"source": [
|
|
"# Correlation: Speech Attributes vs Survey Metrics\n",
|
|
"\n",
|
|
"We correlate each speech characteristic (pitch, duration, jitter, etc.) against two survey metrics:\n",
|
|
"\n",
|
|
"| Metric | Type | Correlation Method | Why |\n",
|
|
"|---|---|---|---|\n",
|
|
"| **Avg Voice Score (1-10)** | Continuous | **Pearson** | Both variables are continuous and approximately interval-scaled — Pearson captures linear relationships well here. |\n",
|
|
"| **Weighted Ranking Score** | Ordinal / count-based | **Spearman** | The weighted score (1st=3pts, 2nd=2pts, 3rd=1pt) is ordinal in nature with a small number of discrete values. Spearman is rank-based, making no assumptions about linearity or normality — more appropriate for this type of data. |\n",
|
|
"\n",
|
|
"> **Note:** With only ~17 voices, all correlations should be interpreted cautiously. Small samples amplify the influence of individual data points."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 41,
|
|
"id": "77658327",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"shape: (39, 3)\n",
|
|
"┌─────────────────────────────────┬─────────────────────────────────┬─────────────┐\n",
|
|
"│ attribute ┆ metric ┆ correlation │\n",
|
|
"│ --- ┆ --- ┆ --- │\n",
|
|
"│ str ┆ str ┆ f64 │\n",
|
|
"╞═════════════════════════════════╪═════════════════════════════════╪═════════════╡\n",
|
|
"│ Age ┆ Avg Voice Score (1-10) [Pearso… ┆ 0.386054 │\n",
|
|
"│ Age ┆ Weighted Ranking Score [Spearm… ┆ 0.094086 │\n",
|
|
"│ Mean pitch ┆ Avg Voice Score (1-10) [Pearso… ┆ 0.459684 │\n",
|
|
"│ Mean pitch ┆ Weighted Ranking Score [Spearm… ┆ 0.63429 │\n",
|
|
"│ Standard deviation pitch ┆ Avg Voice Score (1-10) [Pearso… ┆ 0.664432 │\n",
|
|
"│ … ┆ … ┆ … │\n",
|
|
"│ speechrate (nsyll/dur) ┆ Weighted Ranking Score [Spearm… ┆ -0.336524 │\n",
|
|
"│ articulation rate (nsyll / pho… ┆ Avg Voice Score (1-10) [Pearso… ┆ -0.456181 │\n",
|
|
"│ articulation rate (nsyll / pho… ┆ Weighted Ranking Score [Spearm… ┆ -0.268239 │\n",
|
|
"│ speech rate words per minute ┆ Avg Voice Score (1-10) [Pearso… ┆ -0.26437 │\n",
|
|
"│ speech rate words per minute ┆ Weighted Ranking Score [Spearm… ┆ 0.252577 │\n",
|
|
"└─────────────────────────────────┴─────────────────────────────────┴─────────────┘\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"# --- Compute correlations ---\n",
|
|
"# Pearson for continuous Voice Score, Spearman for ordinal Weighted Ranking Score\n",
|
|
"exclude_cols = {\"Voice\", \"Gender\", \"Avg Voice Score (1-10)\", \"Weighted Score\"}\n",
|
|
"speech_attrs = [c for c in merged.columns if c not in exclude_cols]\n",
|
|
"\n",
|
|
"rows = []\n",
|
|
"for attr in speech_attrs:\n",
|
|
" # Drop nulls for the pair before computing correlation\n",
|
|
" valid = merged.select([attr, \"Avg Voice Score (1-10)\", \"Weighted Score\"]).drop_nulls()\n",
|
|
" if valid.height > 2:\n",
|
|
" # Pearson for continuous 1-10 score\n",
|
|
" r_score = valid.select(pl.corr(attr, \"Avg Voice Score (1-10)\", method=\"pearson\")).item()\n",
|
|
" # Spearman for ordinal weighted ranking score\n",
|
|
" r_rank = valid.select(pl.corr(attr, \"Weighted Score\", method=\"spearman\")).item()\n",
|
|
" else:\n",
|
|
" r_score = None\n",
|
|
" r_rank = None\n",
|
|
" rows.append({\"attribute\": attr, \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": r_score})\n",
|
|
" rows.append({\"attribute\": attr, \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": r_rank})\n",
|
|
"\n",
|
|
"corr_long = (\n",
|
|
" pl.DataFrame(rows)\n",
|
|
" .drop_nulls()\n",
|
|
" .filter(pl.col(\"correlation\").is_not_nan())\n",
|
|
")\n",
|
|
"print(corr_long)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 46,
|
|
"id": "ef4ceefc",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Female Age correlation:\n",
|
|
"shape: (2, 3)\n",
|
|
"┌───────────┬─────────────────────────────────┬─────────────┐\n",
|
|
"│ attribute ┆ metric ┆ correlation │\n",
|
|
"│ --- ┆ --- ┆ --- │\n",
|
|
"│ str ┆ str ┆ f64 │\n",
|
|
"╞═══════════╪═════════════════════════════════╪═════════════╡\n",
|
|
"│ Age ┆ Avg Voice Score (1-10) [Pearso… ┆ -0.023566 │\n",
|
|
"│ Age ┆ Weighted Ranking Score [Spearm… ┆ 0.231908 │\n",
|
|
"└───────────┴─────────────────────────────────┴─────────────┘\n",
|
|
"\n",
|
|
"Female data (Age + Weighted Score):\n",
|
|
"shape: (6, 3)\n",
|
|
"┌───────┬─────┬────────────────┐\n",
|
|
"│ Voice ┆ Age ┆ Weighted Score │\n",
|
|
"│ --- ┆ --- ┆ --- │\n",
|
|
"│ str ┆ i64 ┆ i64 │\n",
|
|
"╞═══════╪═════╪════════════════╡\n",
|
|
"│ V04 ┆ 28 ┆ 209 │\n",
|
|
"│ V89 ┆ 32 ┆ 130 │\n",
|
|
"│ V48 ┆ 35 ┆ 144 │\n",
|
|
"│ V08 ┆ 45 ┆ 180 │\n",
|
|
"│ V77 ┆ 48 ┆ 158 │\n",
|
|
"│ V14 ┆ 50 ┆ 209 │\n",
|
|
"└───────┴─────┴────────────────┘\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"# Verify: Age correlation in female split should no longer be 1.0\n",
|
|
"print(\"Female Age correlation:\")\n",
|
|
"print(corr_female.filter(pl.col(\"attribute\") == \"Age\"))\n",
|
|
"\n",
|
|
"print(\"\\nFemale data (Age + Weighted Score):\")\n",
|
|
"female_check = merged.filter(pl.col(\"Gender\") == \"female\").select([\"Voice\", \"Age\", \"Weighted Score\"]).drop_nulls().sort(\"Age\")\n",
|
|
"print(female_check)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 42,
|
|
"id": "0d9567ff",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Saved plot to figures/2-4-26/All_Respondents/speech_attr_vs_survey_correlation.png\n"
|
|
]
|
|
},
|
|
{
|
|
"data": {
|
|
"text/html": [
|
|
"\n",
|
|
"<style>\n",
|
|
" #altair-viz-edfa78fab61544c2b6298947b6780648.vega-embed {\n",
|
|
" width: 100%;\n",
|
|
" display: flex;\n",
|
|
" }\n",
|
|
"\n",
|
|
" #altair-viz-edfa78fab61544c2b6298947b6780648.vega-embed details,\n",
|
|
" #altair-viz-edfa78fab61544c2b6298947b6780648.vega-embed details summary {\n",
|
|
" position: relative;\n",
|
|
" }\n",
|
|
"</style>\n",
|
|
"<div id=\"altair-viz-edfa78fab61544c2b6298947b6780648\"></div>\n",
|
|
"<script type=\"text/javascript\">\n",
|
|
" var VEGA_DEBUG = (typeof VEGA_DEBUG == \"undefined\") ? {} : VEGA_DEBUG;\n",
|
|
" (function(spec, embedOpt){\n",
|
|
" let outputDiv = document.currentScript.previousElementSibling;\n",
|
|
" if (outputDiv.id !== \"altair-viz-edfa78fab61544c2b6298947b6780648\") {\n",
|
|
" outputDiv = document.getElementById(\"altair-viz-edfa78fab61544c2b6298947b6780648\");\n",
|
|
" }\n",
|
|
"\n",
|
|
" const paths = {\n",
|
|
" \"vega\": \"https://cdn.jsdelivr.net/npm/vega@6?noext\",\n",
|
|
" \"vega-lib\": \"https://cdn.jsdelivr.net/npm/vega-lib?noext\",\n",
|
|
" \"vega-lite\": \"https://cdn.jsdelivr.net/npm/vega-lite@6.1.0?noext\",\n",
|
|
" \"vega-embed\": \"https://cdn.jsdelivr.net/npm/vega-embed@7?noext\",\n",
|
|
" };\n",
|
|
"\n",
|
|
" function maybeLoadScript(lib, version) {\n",
|
|
" var key = `${lib.replace(\"-\", \"\")}_version`;\n",
|
|
" return (VEGA_DEBUG[key] == version) ?\n",
|
|
" Promise.resolve(paths[lib]) :\n",
|
|
" new Promise(function(resolve, reject) {\n",
|
|
" var s = document.createElement('script');\n",
|
|
" document.getElementsByTagName(\"head\")[0].appendChild(s);\n",
|
|
" s.async = true;\n",
|
|
" s.onload = () => {\n",
|
|
" VEGA_DEBUG[key] = version;\n",
|
|
" return resolve(paths[lib]);\n",
|
|
" };\n",
|
|
" s.onerror = () => reject(`Error loading script: ${paths[lib]}`);\n",
|
|
" s.src = paths[lib];\n",
|
|
" });\n",
|
|
" }\n",
|
|
"\n",
|
|
" function showError(err) {\n",
|
|
" outputDiv.innerHTML = `<div class=\"error\" style=\"color:red;\">${err}</div>`;\n",
|
|
" throw err;\n",
|
|
" }\n",
|
|
"\n",
|
|
" function displayChart(vegaEmbed) {\n",
|
|
" vegaEmbed(outputDiv, spec, embedOpt)\n",
|
|
" .catch(err => showError(`Javascript Error: ${err.message}<br>This usually means there's a typo in your chart specification. See the javascript console for the full traceback.`));\n",
|
|
" }\n",
|
|
"\n",
|
|
" if(typeof define === \"function\" && define.amd) {\n",
|
|
" requirejs.config({paths});\n",
|
|
" let deps = [\"vega-embed\"];\n",
|
|
" require(deps, displayChart, err => showError(`Error loading script: ${err.message}`));\n",
|
|
" } else {\n",
|
|
" maybeLoadScript(\"vega\", \"6\")\n",
|
|
" .then(() => maybeLoadScript(\"vega-lite\", \"6.1.0\"))\n",
|
|
" .then(() => maybeLoadScript(\"vega-embed\", \"7\"))\n",
|
|
" .catch(showError)\n",
|
|
" .then(() => displayChart(vegaEmbed));\n",
|
|
" }\n",
|
|
" })({\"config\": {\"view\": {\"continuousWidth\": 1000, \"continuousHeight\": 500, \"strokeWidth\": 0}, \"background\": \"white\", \"axis\": {\"grid\": true, \"gridColor\": \"lightgray\", \"labelFontSize\": 11, \"titleFontSize\": 12, \"labelColor\": \"black\", \"titleColor\": \"black\", \"labelLimit\": 200}, \"axisX\": {\"labelAngle\": -45, \"labelLimit\": 200}, \"axisY\": {\"labelAngle\": 0}, \"legend\": {\"orient\": \"top\", \"direction\": \"horizontal\", \"titleFontSize\": 11, \"labelFontSize\": 11}, \"title\": {\"fontSize\": 14, \"color\": \"black\", \"anchor\": \"start\", \"subtitleFontSize\": 10, \"subtitleColor\": \"gray\"}, \"bar\": {\"color\": \"#0077B6\"}}, \"layer\": [{\"data\": {\"name\": \"data-bf2d69abcead9f87074f0225263a6b70\"}, \"mark\": {\"type\": \"rect\", \"stroke\": \"white\", \"strokeWidth\": 1}, \"encoding\": {\"color\": {\"field\": \"correlation\", \"legend\": {\"title\": \"Pearson r\"}, \"scale\": {\"domain\": [-1, 1], \"scheme\": \"redblue\"}, \"type\": \"quantitative\"}, \"tooltip\": [{\"field\": \"metric\", \"title\": \"Metric\", \"type\": \"nominal\"}, {\"field\": \"attribute\", \"title\": \"Attribute\", \"type\": \"nominal\"}, {\"field\": \"correlation\", \"format\": \".3f\", \"title\": \"r\", \"type\": \"quantitative\"}], \"x\": {\"axis\": {\"grid\": false, \"labelAngle\": -45, \"labelLimit\": 180}, \"field\": \"attribute\", \"sort\": [\"Age\", \"Mean pitch\", \"Standard deviation pitch\", \"Minimum pitch\", \"Maximum pitch\", \"Number of voice breaks\", \"Jitter (local)\", \"Shimmer (local)\", \"Mean harmonics-to-noise ratio dB\", \"H1-minus-H2\", \"MinI(dB)\", \"MaxI(dB)\", \"MeanI(dB)\", \"SDI(dB)\", \"MeanAbsSlope(dB/s)\", \"dur (s)\", \"phonationtime (s)\", \"speechrate (nsyll/dur)\", \"articulation rate (nsyll / phonationtime)\", \"speech rate words per minute\"], \"title\": null, \"type\": \"nominal\"}, \"y\": {\"axis\": {\"grid\": false, \"labelLimit\": 200}, \"field\": \"metric\", \"sort\": [\"Avg Voice Score (1-10) [Pearson]\", \"Weighted Ranking Score [Spearman]\"], \"title\": null, \"type\": \"nominal\"}}}, {\"data\": {\"name\": \"data-289f8b6b88c4bd0d6c4cd2d7944b9204\"}, \"mark\": {\"type\": \"text\", \"color\": \"black\", \"fontSize\": 11, \"fontWeight\": \"normal\"}, \"encoding\": {\"text\": {\"field\": \"correlation\", \"format\": \".2f\", \"type\": \"quantitative\"}, \"x\": {\"field\": \"attribute\", \"sort\": [\"Age\", \"Mean pitch\", \"Standard deviation pitch\", \"Minimum pitch\", \"Maximum pitch\", \"Number of voice breaks\", \"Jitter (local)\", \"Shimmer (local)\", \"Mean harmonics-to-noise ratio dB\", \"H1-minus-H2\", \"MinI(dB)\", \"MaxI(dB)\", \"MeanI(dB)\", \"SDI(dB)\", \"MeanAbsSlope(dB/s)\", \"dur (s)\", \"phonationtime (s)\", \"speechrate (nsyll/dur)\", \"articulation rate (nsyll / phonationtime)\", \"speech rate words per minute\"], \"type\": \"nominal\"}, \"y\": {\"field\": \"metric\", \"sort\": [\"Avg Voice Score (1-10) [Pearson]\", \"Weighted Ranking Score [Spearman]\"], \"type\": \"nominal\"}}}, {\"data\": {\"name\": \"data-d9febc5cd73fea8c5b239d0e0fcc4730\"}, \"mark\": {\"type\": \"text\", \"color\": \"white\", \"fontSize\": 11, \"fontWeight\": \"normal\"}, \"encoding\": {\"text\": {\"field\": \"correlation\", \"format\": \".2f\", \"type\": \"quantitative\"}, \"x\": {\"field\": \"attribute\", \"sort\": [\"Age\", \"Mean pitch\", \"Standard deviation pitch\", \"Minimum pitch\", \"Maximum pitch\", \"Number of voice breaks\", \"Jitter (local)\", \"Shimmer (local)\", \"Mean harmonics-to-noise ratio dB\", \"H1-minus-H2\", \"MinI(dB)\", \"MaxI(dB)\", \"MeanI(dB)\", \"SDI(dB)\", \"MeanAbsSlope(dB/s)\", \"dur (s)\", \"phonationtime (s)\", \"speechrate (nsyll/dur)\", \"articulation rate (nsyll / phonationtime)\", \"speech rate words per minute\"], \"type\": \"nominal\"}, \"y\": {\"field\": \"metric\", \"sort\": [\"Avg Voice Score (1-10) [Pearson]\", \"Weighted Ranking Score [Spearman]\"], \"type\": \"nominal\"}}}], \"height\": 160, \"title\": {\"text\": [\"Speech Characteristics vs Survey Metrics\", \"Correlation per Voice (Pearson / Spearman)\"], \"subtitle\": [\"Sample size: 455\"], \"subtitleColor\": \"gray\", \"subtitleFontSize\": 10, \"anchor\": \"start\"}, \"width\": 1100, \"$schema\": \"https://vega.github.io/schema/vega-lite/v6.1.0.json\", \"datasets\": {\"data-bf2d69abcead9f87074f0225263a6b70\": [{\"attribute\": \"Age\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.3860542538211813}, {\"attribute\": \"Age\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.09408617259260567}, {\"attribute\": \"Mean pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.4596840564490561}, {\"attribute\": \"Mean pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.6342900907774179}, {\"attribute\": \"Standard deviation pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.6644317974322577}, {\"attribute\": \"Standard deviation pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.4621046136986293}, {\"attribute\": \"Minimum pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.1759369593953851}, {\"attribute\": \"Minimum pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.31052252703952476}, {\"attribute\": \"Maximum pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.5017986100730272}, {\"attribute\": \"Maximum pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.12950702549515725}, {\"attribute\": \"Number of voice breaks\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.6245665417092515}, {\"attribute\": \"Number of voice breaks\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.4793226233717229}, {\"attribute\": \"Jitter (local)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.21256639799177363}, {\"attribute\": \"Jitter (local)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.5552282768777614}, {\"attribute\": \"Shimmer (local)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.06278597743642035}, {\"attribute\": \"Shimmer (local)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.47387797965273454}, {\"attribute\": \"Mean harmonics-to-noise ratio dB\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.14672115371769612}, {\"attribute\": \"Mean harmonics-to-noise ratio dB\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.478292991885524}, {\"attribute\": \"H1-minus-H2\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.2632790009685734}, {\"attribute\": \"H1-minus-H2\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.5974983221708392}, {\"attribute\": \"MinI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.13144303593482204}, {\"attribute\": \"MinI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.026509572901325478}, {\"attribute\": \"MaxI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.08205229565874973}, {\"attribute\": \"MaxI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.20014722121978848}, {\"attribute\": \"MeanI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.2313204819112786}, {\"attribute\": \"MeanI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.3949890534544011}, {\"attribute\": \"SDI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.287746169841669}, {\"attribute\": \"SDI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.0}, {\"attribute\": \"MeanAbsSlope(dB/s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.3573794459254656}, {\"attribute\": \"MeanAbsSlope(dB/s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.36203100308873504}, {\"attribute\": \"dur (s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.25521688716065966}, {\"attribute\": \"dur (s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.25257731958762886}, {\"attribute\": \"phonationtime (s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.4556993131527357}, {\"attribute\": \"phonationtime (s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.26823883480858585}, {\"attribute\": \"speechrate (nsyll/dur)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.33652430044182624}, {\"attribute\": \"articulation rate (nsyll / phonationtime)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.45618077972667986}, {\"attribute\": \"articulation rate (nsyll / phonationtime)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.26823883480858585}, {\"attribute\": \"speech rate words per minute\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.26437048461934665}, {\"attribute\": \"speech rate words per minute\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.25257731958762886}], \"data-289f8b6b88c4bd0d6c4cd2d7944b9204\": [{\"attribute\": \"Age\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.3860542538211813}, {\"attribute\": \"Age\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.09408617259260567}, {\"attribute\": \"Minimum pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.1759369593953851}, {\"attribute\": \"Minimum pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.31052252703952476}, {\"attribute\": \"Maximum pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.12950702549515725}, {\"attribute\": \"Jitter (local)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.21256639799177363}, {\"attribute\": \"Shimmer (local)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.06278597743642035}, {\"attribute\": \"Mean harmonics-to-noise ratio dB\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.14672115371769612}, {\"attribute\": \"H1-minus-H2\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.2632790009685734}, {\"attribute\": \"MinI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.13144303593482204}, {\"attribute\": \"MinI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.026509572901325478}, {\"attribute\": \"MaxI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.08205229565874973}, {\"attribute\": \"MaxI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.20014722121978848}, {\"attribute\": \"MeanI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.2313204819112786}, {\"attribute\": \"MeanI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.3949890534544011}, {\"attribute\": \"SDI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.287746169841669}, {\"attribute\": \"SDI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.0}, {\"attribute\": \"MeanAbsSlope(dB/s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.3573794459254656}, {\"attribute\": \"MeanAbsSlope(dB/s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.36203100308873504}, {\"attribute\": \"dur (s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.25521688716065966}, {\"attribute\": \"dur (s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.25257731958762886}, {\"attribute\": \"phonationtime (s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.26823883480858585}, {\"attribute\": \"speechrate (nsyll/dur)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.33652430044182624}, {\"attribute\": \"articulation rate (nsyll / phonationtime)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.26823883480858585}, {\"attribute\": \"speech rate words per minute\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.26437048461934665}, {\"attribute\": \"speech rate words per minute\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.25257731958762886}], \"data-d9febc5cd73fea8c5b239d0e0fcc4730\": [{\"attribute\": \"Mean pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.4596840564490561}, {\"attribute\": \"Mean pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.6342900907774179}, {\"attribute\": \"Standard deviation pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.6644317974322577}, {\"attribute\": \"Standard deviation pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.4621046136986293}, {\"attribute\": \"Maximum pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.5017986100730272}, {\"attribute\": \"Number of voice breaks\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.6245665417092515}, {\"attribute\": \"Number of voice breaks\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.4793226233717229}, {\"attribute\": \"Jitter (local)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.5552282768777614}, {\"attribute\": \"Shimmer (local)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.47387797965273454}, {\"attribute\": \"Mean harmonics-to-noise ratio dB\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.478292991885524}, {\"attribute\": \"H1-minus-H2\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.5974983221708392}, {\"attribute\": \"phonationtime (s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.4556993131527357}, {\"attribute\": \"articulation rate (nsyll / phonationtime)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.45618077972667986}]}}, {\"mode\": \"vega-lite\"});\n",
|
|
"</script>"
|
|
],
|
|
"text/plain": [
|
|
"alt.LayerChart(...)"
|
|
]
|
|
},
|
|
"execution_count": 42,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"# --- Plot correlation heatmap ---\n",
|
|
"S.plot_speech_attribute_correlation(\n",
|
|
" corr_long,\n",
|
|
" title=\"Speech Characteristics vs Survey Metrics<br>Correlation per Voice (Pearson / Spearman)\",\n",
|
|
" filename=\"speech_attr_vs_survey_correlation\",\n",
|
|
")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 43,
|
|
"id": "a173be9a",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Male voices: 10\n",
|
|
"Female voices: 8\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"# --- Compute correlations by voice gender ---\n",
|
|
"def compute_corr_for_gender(merged_df: pl.DataFrame, gender: str) -> pl.DataFrame:\n",
|
|
" \"\"\"Compute Pearson (score) + Spearman (ranking) correlations for a gender subset.\"\"\"\n",
|
|
" subset = merged_df.filter(pl.col(\"Gender\") == gender)\n",
|
|
" exclude = {\"Voice\", \"Gender\", \"Avg Voice Score (1-10)\", \"Weighted Score\"}\n",
|
|
" attrs = [c for c in subset.columns if c not in exclude]\n",
|
|
"\n",
|
|
" rows = []\n",
|
|
" for attr in attrs:\n",
|
|
" valid = subset.select([attr, \"Avg Voice Score (1-10)\", \"Weighted Score\"]).drop_nulls()\n",
|
|
" if valid.height > 2:\n",
|
|
" r_score = valid.select(pl.corr(attr, \"Avg Voice Score (1-10)\", method=\"pearson\")).item()\n",
|
|
" r_rank = valid.select(pl.corr(attr, \"Weighted Score\", method=\"spearman\")).item()\n",
|
|
" else:\n",
|
|
" r_score = None\n",
|
|
" r_rank = None\n",
|
|
" rows.append({\"attribute\": attr, \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": r_score})\n",
|
|
" rows.append({\"attribute\": attr, \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": r_rank})\n",
|
|
"\n",
|
|
" return pl.DataFrame(rows).drop_nulls().filter(pl.col(\"correlation\").is_not_nan())\n",
|
|
"\n",
|
|
"corr_male = compute_corr_for_gender(merged, \"male\")\n",
|
|
"corr_female = compute_corr_for_gender(merged, \"female\")\n",
|
|
"\n",
|
|
"print(f\"Male voices: {merged.filter(pl.col('Gender') == 'male').height}\")\n",
|
|
"print(f\"Female voices: {merged.filter(pl.col('Gender') == 'female').height}\")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 44,
|
|
"id": "84eaaff6",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Saved plot to figures/2-4-26/All_Respondents/speech_attr_vs_survey_correlation_male.png\n"
|
|
]
|
|
},
|
|
{
|
|
"data": {
|
|
"text/html": [
|
|
"\n",
|
|
"<style>\n",
|
|
" #altair-viz-8bb6e49fc5024b13bd65cd65e7abcc26.vega-embed {\n",
|
|
" width: 100%;\n",
|
|
" display: flex;\n",
|
|
" }\n",
|
|
"\n",
|
|
" #altair-viz-8bb6e49fc5024b13bd65cd65e7abcc26.vega-embed details,\n",
|
|
" #altair-viz-8bb6e49fc5024b13bd65cd65e7abcc26.vega-embed details summary {\n",
|
|
" position: relative;\n",
|
|
" }\n",
|
|
"</style>\n",
|
|
"<div id=\"altair-viz-8bb6e49fc5024b13bd65cd65e7abcc26\"></div>\n",
|
|
"<script type=\"text/javascript\">\n",
|
|
" var VEGA_DEBUG = (typeof VEGA_DEBUG == \"undefined\") ? {} : VEGA_DEBUG;\n",
|
|
" (function(spec, embedOpt){\n",
|
|
" let outputDiv = document.currentScript.previousElementSibling;\n",
|
|
" if (outputDiv.id !== \"altair-viz-8bb6e49fc5024b13bd65cd65e7abcc26\") {\n",
|
|
" outputDiv = document.getElementById(\"altair-viz-8bb6e49fc5024b13bd65cd65e7abcc26\");\n",
|
|
" }\n",
|
|
"\n",
|
|
" const paths = {\n",
|
|
" \"vega\": \"https://cdn.jsdelivr.net/npm/vega@6?noext\",\n",
|
|
" \"vega-lib\": \"https://cdn.jsdelivr.net/npm/vega-lib?noext\",\n",
|
|
" \"vega-lite\": \"https://cdn.jsdelivr.net/npm/vega-lite@6.1.0?noext\",\n",
|
|
" \"vega-embed\": \"https://cdn.jsdelivr.net/npm/vega-embed@7?noext\",\n",
|
|
" };\n",
|
|
"\n",
|
|
" function maybeLoadScript(lib, version) {\n",
|
|
" var key = `${lib.replace(\"-\", \"\")}_version`;\n",
|
|
" return (VEGA_DEBUG[key] == version) ?\n",
|
|
" Promise.resolve(paths[lib]) :\n",
|
|
" new Promise(function(resolve, reject) {\n",
|
|
" var s = document.createElement('script');\n",
|
|
" document.getElementsByTagName(\"head\")[0].appendChild(s);\n",
|
|
" s.async = true;\n",
|
|
" s.onload = () => {\n",
|
|
" VEGA_DEBUG[key] = version;\n",
|
|
" return resolve(paths[lib]);\n",
|
|
" };\n",
|
|
" s.onerror = () => reject(`Error loading script: ${paths[lib]}`);\n",
|
|
" s.src = paths[lib];\n",
|
|
" });\n",
|
|
" }\n",
|
|
"\n",
|
|
" function showError(err) {\n",
|
|
" outputDiv.innerHTML = `<div class=\"error\" style=\"color:red;\">${err}</div>`;\n",
|
|
" throw err;\n",
|
|
" }\n",
|
|
"\n",
|
|
" function displayChart(vegaEmbed) {\n",
|
|
" vegaEmbed(outputDiv, spec, embedOpt)\n",
|
|
" .catch(err => showError(`Javascript Error: ${err.message}<br>This usually means there's a typo in your chart specification. See the javascript console for the full traceback.`));\n",
|
|
" }\n",
|
|
"\n",
|
|
" if(typeof define === \"function\" && define.amd) {\n",
|
|
" requirejs.config({paths});\n",
|
|
" let deps = [\"vega-embed\"];\n",
|
|
" require(deps, displayChart, err => showError(`Error loading script: ${err.message}`));\n",
|
|
" } else {\n",
|
|
" maybeLoadScript(\"vega\", \"6\")\n",
|
|
" .then(() => maybeLoadScript(\"vega-lite\", \"6.1.0\"))\n",
|
|
" .then(() => maybeLoadScript(\"vega-embed\", \"7\"))\n",
|
|
" .catch(showError)\n",
|
|
" .then(() => displayChart(vegaEmbed));\n",
|
|
" }\n",
|
|
" })({\"config\": {\"view\": {\"continuousWidth\": 1000, \"continuousHeight\": 500, \"strokeWidth\": 0}, \"background\": \"white\", \"axis\": {\"grid\": true, \"gridColor\": \"lightgray\", \"labelFontSize\": 11, \"titleFontSize\": 12, \"labelColor\": \"black\", \"titleColor\": \"black\", \"labelLimit\": 200}, \"axisX\": {\"labelAngle\": -45, \"labelLimit\": 200}, \"axisY\": {\"labelAngle\": 0}, \"legend\": {\"orient\": \"top\", \"direction\": \"horizontal\", \"titleFontSize\": 11, \"labelFontSize\": 11}, \"title\": {\"fontSize\": 14, \"color\": \"black\", \"anchor\": \"start\", \"subtitleFontSize\": 10, \"subtitleColor\": \"gray\"}, \"bar\": {\"color\": \"#0077B6\"}}, \"layer\": [{\"data\": {\"name\": \"data-294148e4f8046232fd1444ad2a71061b\"}, \"mark\": {\"type\": \"rect\", \"stroke\": \"white\", \"strokeWidth\": 1}, \"encoding\": {\"color\": {\"field\": \"correlation\", \"legend\": {\"title\": \"Pearson r\"}, \"scale\": {\"domain\": [-1, 1], \"scheme\": \"redblue\"}, \"type\": \"quantitative\"}, \"tooltip\": [{\"field\": \"metric\", \"title\": \"Metric\", \"type\": \"nominal\"}, {\"field\": \"attribute\", \"title\": \"Attribute\", \"type\": \"nominal\"}, {\"field\": \"correlation\", \"format\": \".3f\", \"title\": \"r\", \"type\": \"quantitative\"}], \"x\": {\"axis\": {\"grid\": false, \"labelAngle\": -45, \"labelLimit\": 180}, \"field\": \"attribute\", \"sort\": [\"Age\", \"Mean pitch\", \"Standard deviation pitch\", \"Minimum pitch\", \"Maximum pitch\", \"Number of voice breaks\", \"Jitter (local)\", \"Shimmer (local)\", \"Mean harmonics-to-noise ratio dB\", \"H1-minus-H2\", \"MinI(dB)\", \"MaxI(dB)\", \"MeanI(dB)\", \"SDI(dB)\", \"MeanAbsSlope(dB/s)\", \"dur (s)\", \"phonationtime (s)\", \"speechrate (nsyll/dur)\", \"articulation rate (nsyll / phonationtime)\", \"speech rate words per minute\"], \"title\": null, \"type\": \"nominal\"}, \"y\": {\"axis\": {\"grid\": false, \"labelLimit\": 200}, \"field\": \"metric\", \"sort\": [\"Avg Voice Score (1-10) [Pearson]\", \"Weighted Ranking Score [Spearman]\"], \"title\": null, \"type\": \"nominal\"}}}, {\"data\": {\"name\": \"data-831ec53dcaaeecdb39d6b0d9725a3191\"}, \"mark\": {\"type\": \"text\", \"color\": \"black\", \"fontSize\": 11, \"fontWeight\": \"normal\"}, \"encoding\": {\"text\": {\"field\": \"correlation\", \"format\": \".2f\", \"type\": \"quantitative\"}, \"x\": {\"field\": \"attribute\", \"sort\": [\"Age\", \"Mean pitch\", \"Standard deviation pitch\", \"Minimum pitch\", \"Maximum pitch\", \"Number of voice breaks\", \"Jitter (local)\", \"Shimmer (local)\", \"Mean harmonics-to-noise ratio dB\", \"H1-minus-H2\", \"MinI(dB)\", \"MaxI(dB)\", \"MeanI(dB)\", \"SDI(dB)\", \"MeanAbsSlope(dB/s)\", \"dur (s)\", \"phonationtime (s)\", \"speechrate (nsyll/dur)\", \"articulation rate (nsyll / phonationtime)\", \"speech rate words per minute\"], \"type\": \"nominal\"}, \"y\": {\"field\": \"metric\", \"sort\": [\"Avg Voice Score (1-10) [Pearson]\", \"Weighted Ranking Score [Spearman]\"], \"type\": \"nominal\"}}}, {\"data\": {\"name\": \"data-52bf66d21049bc948cf856353572dbed\"}, \"mark\": {\"type\": \"text\", \"color\": \"white\", \"fontSize\": 11, \"fontWeight\": \"normal\"}, \"encoding\": {\"text\": {\"field\": \"correlation\", \"format\": \".2f\", \"type\": \"quantitative\"}, \"x\": {\"field\": \"attribute\", \"sort\": [\"Age\", \"Mean pitch\", \"Standard deviation pitch\", \"Minimum pitch\", \"Maximum pitch\", \"Number of voice breaks\", \"Jitter (local)\", \"Shimmer (local)\", \"Mean harmonics-to-noise ratio dB\", \"H1-minus-H2\", \"MinI(dB)\", \"MaxI(dB)\", \"MeanI(dB)\", \"SDI(dB)\", \"MeanAbsSlope(dB/s)\", \"dur (s)\", \"phonationtime (s)\", \"speechrate (nsyll/dur)\", \"articulation rate (nsyll / phonationtime)\", \"speech rate words per minute\"], \"type\": \"nominal\"}, \"y\": {\"field\": \"metric\", \"sort\": [\"Avg Voice Score (1-10) [Pearson]\", \"Weighted Ranking Score [Spearman]\"], \"type\": \"nominal\"}}}], \"height\": 160, \"title\": {\"text\": [\"Speech Characteristics vs Survey Metrics\", \"Male Voices Only (Pearson / Spearman)\"], \"subtitle\": [\"Sample size: 455\"], \"subtitleColor\": \"gray\", \"subtitleFontSize\": 10, \"anchor\": \"start\"}, \"width\": 1100, \"$schema\": \"https://vega.github.io/schema/vega-lite/v6.1.0.json\", \"datasets\": {\"data-294148e4f8046232fd1444ad2a71061b\": [{\"attribute\": \"Age\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.7927291691160427}, {\"attribute\": \"Age\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.11765121295424251}, {\"attribute\": \"Mean pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.07387665732270982}, {\"attribute\": \"Mean pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.31666666666666665}, {\"attribute\": \"Standard deviation pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.6034368859027126}, {\"attribute\": \"Standard deviation pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.016666666666666666}, {\"attribute\": \"Minimum pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.045593534007381364}, {\"attribute\": \"Minimum pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.5333333333333333}, {\"attribute\": \"Maximum pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.6204407338182272}, {\"attribute\": \"Maximum pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.21666666666666667}, {\"attribute\": \"Number of voice breaks\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.3800450521306157}, {\"attribute\": \"Number of voice breaks\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.10084389681792215}, {\"attribute\": \"Jitter (local)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.31567445527294263}, {\"attribute\": \"Jitter (local)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.13333333333333333}, {\"attribute\": \"Shimmer (local)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.8125424009949568}, {\"attribute\": \"Shimmer (local)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.36666666666666664}, {\"attribute\": \"Mean harmonics-to-noise ratio dB\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.7383607657633772}, {\"attribute\": \"Mean harmonics-to-noise ratio dB\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.7666666666666667}, {\"attribute\": \"H1-minus-H2\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.11917236523792353}, {\"attribute\": \"H1-minus-H2\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.3}, {\"attribute\": \"MinI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.3263747881906577}, {\"attribute\": \"MinI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.2}, {\"attribute\": \"MaxI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.06717033815227702}, {\"attribute\": \"MaxI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.13333333333333333}, {\"attribute\": \"MeanI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.031002439933730502}, {\"attribute\": \"MeanI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.2928895930645031}, {\"attribute\": \"SDI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.5633028462359824}, {\"attribute\": \"SDI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.26666666666666666}, {\"attribute\": \"MeanAbsSlope(dB/s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.4005768082277356}, {\"attribute\": \"MeanAbsSlope(dB/s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.23333333333333334}, {\"attribute\": \"dur (s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.4609220168794499}, {\"attribute\": \"dur (s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.5}, {\"attribute\": \"phonationtime (s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.4065438796132842}, {\"attribute\": \"phonationtime (s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.26666666666666666}, {\"attribute\": \"speechrate (nsyll/dur)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.4}, {\"attribute\": \"articulation rate (nsyll / phonationtime)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.39263591023580885}, {\"attribute\": \"articulation rate (nsyll / phonationtime)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.26666666666666666}, {\"attribute\": \"speech rate words per minute\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.46339784895869474}, {\"attribute\": \"speech rate words per minute\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.5}], \"data-831ec53dcaaeecdb39d6b0d9725a3191\": [{\"attribute\": \"Age\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.11765121295424251}, {\"attribute\": \"Mean pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.07387665732270982}, {\"attribute\": \"Mean pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.31666666666666665}, {\"attribute\": \"Standard deviation pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.016666666666666666}, {\"attribute\": \"Minimum pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.045593534007381364}, {\"attribute\": \"Maximum pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.21666666666666667}, {\"attribute\": \"Number of voice breaks\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.3800450521306157}, {\"attribute\": \"Number of voice breaks\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.10084389681792215}, {\"attribute\": \"Jitter (local)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.31567445527294263}, {\"attribute\": \"Jitter (local)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.13333333333333333}, {\"attribute\": \"Shimmer (local)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.36666666666666664}, {\"attribute\": \"H1-minus-H2\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.11917236523792353}, {\"attribute\": \"H1-minus-H2\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.3}, {\"attribute\": \"MinI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.3263747881906577}, {\"attribute\": \"MinI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.2}, {\"attribute\": \"MaxI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.06717033815227702}, {\"attribute\": \"MaxI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.13333333333333333}, {\"attribute\": \"MeanI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.031002439933730502}, {\"attribute\": \"MeanI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.2928895930645031}, {\"attribute\": \"SDI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.26666666666666666}, {\"attribute\": \"MeanAbsSlope(dB/s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.4005768082277356}, {\"attribute\": \"MeanAbsSlope(dB/s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.23333333333333334}, {\"attribute\": \"phonationtime (s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.4065438796132842}, {\"attribute\": \"phonationtime (s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.26666666666666666}, {\"attribute\": \"speechrate (nsyll/dur)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.4}, {\"attribute\": \"articulation rate (nsyll / phonationtime)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.39263591023580885}, {\"attribute\": \"articulation rate (nsyll / phonationtime)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.26666666666666666}], \"data-52bf66d21049bc948cf856353572dbed\": [{\"attribute\": \"Age\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.7927291691160427}, {\"attribute\": \"Standard deviation pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.6034368859027126}, {\"attribute\": \"Minimum pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.5333333333333333}, {\"attribute\": \"Maximum pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.6204407338182272}, {\"attribute\": \"Shimmer (local)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.8125424009949568}, {\"attribute\": \"Mean harmonics-to-noise ratio dB\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.7383607657633772}, {\"attribute\": \"Mean harmonics-to-noise ratio dB\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.7666666666666667}, {\"attribute\": \"SDI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.5633028462359824}, {\"attribute\": \"dur (s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.4609220168794499}, {\"attribute\": \"dur (s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.5}, {\"attribute\": \"speech rate words per minute\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.46339784895869474}, {\"attribute\": \"speech rate words per minute\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.5}]}}, {\"mode\": \"vega-lite\"});\n",
|
|
"</script>"
|
|
],
|
|
"text/plain": [
|
|
"alt.LayerChart(...)"
|
|
]
|
|
},
|
|
"execution_count": 44,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"# --- Plot: Male voices ---\n",
|
|
"S.plot_speech_attribute_correlation(\n",
|
|
" corr_male,\n",
|
|
" title=\"Speech Characteristics vs Survey Metrics<br>Male Voices Only (Pearson / Spearman)\",\n",
|
|
" filename=\"speech_attr_vs_survey_correlation_male\",\n",
|
|
")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 45,
|
|
"id": "d04225e1",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Saved plot to figures/2-4-26/All_Respondents/speech_attr_vs_survey_correlation_female.png\n"
|
|
]
|
|
},
|
|
{
|
|
"data": {
|
|
"text/html": [
|
|
"\n",
|
|
"<style>\n",
|
|
" #altair-viz-1dc66c91a21f4b8d9e49612f81eebd9a.vega-embed {\n",
|
|
" width: 100%;\n",
|
|
" display: flex;\n",
|
|
" }\n",
|
|
"\n",
|
|
" #altair-viz-1dc66c91a21f4b8d9e49612f81eebd9a.vega-embed details,\n",
|
|
" #altair-viz-1dc66c91a21f4b8d9e49612f81eebd9a.vega-embed details summary {\n",
|
|
" position: relative;\n",
|
|
" }\n",
|
|
"</style>\n",
|
|
"<div id=\"altair-viz-1dc66c91a21f4b8d9e49612f81eebd9a\"></div>\n",
|
|
"<script type=\"text/javascript\">\n",
|
|
" var VEGA_DEBUG = (typeof VEGA_DEBUG == \"undefined\") ? {} : VEGA_DEBUG;\n",
|
|
" (function(spec, embedOpt){\n",
|
|
" let outputDiv = document.currentScript.previousElementSibling;\n",
|
|
" if (outputDiv.id !== \"altair-viz-1dc66c91a21f4b8d9e49612f81eebd9a\") {\n",
|
|
" outputDiv = document.getElementById(\"altair-viz-1dc66c91a21f4b8d9e49612f81eebd9a\");\n",
|
|
" }\n",
|
|
"\n",
|
|
" const paths = {\n",
|
|
" \"vega\": \"https://cdn.jsdelivr.net/npm/vega@6?noext\",\n",
|
|
" \"vega-lib\": \"https://cdn.jsdelivr.net/npm/vega-lib?noext\",\n",
|
|
" \"vega-lite\": \"https://cdn.jsdelivr.net/npm/vega-lite@6.1.0?noext\",\n",
|
|
" \"vega-embed\": \"https://cdn.jsdelivr.net/npm/vega-embed@7?noext\",\n",
|
|
" };\n",
|
|
"\n",
|
|
" function maybeLoadScript(lib, version) {\n",
|
|
" var key = `${lib.replace(\"-\", \"\")}_version`;\n",
|
|
" return (VEGA_DEBUG[key] == version) ?\n",
|
|
" Promise.resolve(paths[lib]) :\n",
|
|
" new Promise(function(resolve, reject) {\n",
|
|
" var s = document.createElement('script');\n",
|
|
" document.getElementsByTagName(\"head\")[0].appendChild(s);\n",
|
|
" s.async = true;\n",
|
|
" s.onload = () => {\n",
|
|
" VEGA_DEBUG[key] = version;\n",
|
|
" return resolve(paths[lib]);\n",
|
|
" };\n",
|
|
" s.onerror = () => reject(`Error loading script: ${paths[lib]}`);\n",
|
|
" s.src = paths[lib];\n",
|
|
" });\n",
|
|
" }\n",
|
|
"\n",
|
|
" function showError(err) {\n",
|
|
" outputDiv.innerHTML = `<div class=\"error\" style=\"color:red;\">${err}</div>`;\n",
|
|
" throw err;\n",
|
|
" }\n",
|
|
"\n",
|
|
" function displayChart(vegaEmbed) {\n",
|
|
" vegaEmbed(outputDiv, spec, embedOpt)\n",
|
|
" .catch(err => showError(`Javascript Error: ${err.message}<br>This usually means there's a typo in your chart specification. See the javascript console for the full traceback.`));\n",
|
|
" }\n",
|
|
"\n",
|
|
" if(typeof define === \"function\" && define.amd) {\n",
|
|
" requirejs.config({paths});\n",
|
|
" let deps = [\"vega-embed\"];\n",
|
|
" require(deps, displayChart, err => showError(`Error loading script: ${err.message}`));\n",
|
|
" } else {\n",
|
|
" maybeLoadScript(\"vega\", \"6\")\n",
|
|
" .then(() => maybeLoadScript(\"vega-lite\", \"6.1.0\"))\n",
|
|
" .then(() => maybeLoadScript(\"vega-embed\", \"7\"))\n",
|
|
" .catch(showError)\n",
|
|
" .then(() => displayChart(vegaEmbed));\n",
|
|
" }\n",
|
|
" })({\"config\": {\"view\": {\"continuousWidth\": 1000, \"continuousHeight\": 500, \"strokeWidth\": 0}, \"background\": \"white\", \"axis\": {\"grid\": true, \"gridColor\": \"lightgray\", \"labelFontSize\": 11, \"titleFontSize\": 12, \"labelColor\": \"black\", \"titleColor\": \"black\", \"labelLimit\": 200}, \"axisX\": {\"labelAngle\": -45, \"labelLimit\": 200}, \"axisY\": {\"labelAngle\": 0}, \"legend\": {\"orient\": \"top\", \"direction\": \"horizontal\", \"titleFontSize\": 11, \"labelFontSize\": 11}, \"title\": {\"fontSize\": 14, \"color\": \"black\", \"anchor\": \"start\", \"subtitleFontSize\": 10, \"subtitleColor\": \"gray\"}, \"bar\": {\"color\": \"#0077B6\"}}, \"layer\": [{\"data\": {\"name\": \"data-ff3d93a394a88d44a18ee07d1bece27c\"}, \"mark\": {\"type\": \"rect\", \"stroke\": \"white\", \"strokeWidth\": 1}, \"encoding\": {\"color\": {\"field\": \"correlation\", \"legend\": {\"title\": \"Pearson r\"}, \"scale\": {\"domain\": [-1, 1], \"scheme\": \"redblue\"}, \"type\": \"quantitative\"}, \"tooltip\": [{\"field\": \"metric\", \"title\": \"Metric\", \"type\": \"nominal\"}, {\"field\": \"attribute\", \"title\": \"Attribute\", \"type\": \"nominal\"}, {\"field\": \"correlation\", \"format\": \".3f\", \"title\": \"r\", \"type\": \"quantitative\"}], \"x\": {\"axis\": {\"grid\": false, \"labelAngle\": -45, \"labelLimit\": 180}, \"field\": \"attribute\", \"sort\": [\"Age\", \"Mean pitch\", \"Standard deviation pitch\", \"Minimum pitch\", \"Maximum pitch\", \"Number of voice breaks\", \"Jitter (local)\", \"Shimmer (local)\", \"Mean harmonics-to-noise ratio dB\", \"H1-minus-H2\", \"MinI(dB)\", \"MaxI(dB)\", \"MeanI(dB)\", \"SDI(dB)\", \"MeanAbsSlope(dB/s)\", \"dur (s)\", \"phonationtime (s)\", \"speechrate (nsyll/dur)\", \"articulation rate (nsyll / phonationtime)\", \"speech rate words per minute\"], \"title\": null, \"type\": \"nominal\"}, \"y\": {\"axis\": {\"grid\": false, \"labelLimit\": 200}, \"field\": \"metric\", \"sort\": [\"Avg Voice Score (1-10) [Pearson]\", \"Weighted Ranking Score [Spearman]\"], \"title\": null, \"type\": \"nominal\"}}}, {\"data\": {\"name\": \"data-46f3f309a17840278531f68da6ea94e1\"}, \"mark\": {\"type\": \"text\", \"color\": \"black\", \"fontSize\": 11, \"fontWeight\": \"normal\"}, \"encoding\": {\"text\": {\"field\": \"correlation\", \"format\": \".2f\", \"type\": \"quantitative\"}, \"x\": {\"field\": \"attribute\", \"sort\": [\"Age\", \"Mean pitch\", \"Standard deviation pitch\", \"Minimum pitch\", \"Maximum pitch\", \"Number of voice breaks\", \"Jitter (local)\", \"Shimmer (local)\", \"Mean harmonics-to-noise ratio dB\", \"H1-minus-H2\", \"MinI(dB)\", \"MaxI(dB)\", \"MeanI(dB)\", \"SDI(dB)\", \"MeanAbsSlope(dB/s)\", \"dur (s)\", \"phonationtime (s)\", \"speechrate (nsyll/dur)\", \"articulation rate (nsyll / phonationtime)\", \"speech rate words per minute\"], \"type\": \"nominal\"}, \"y\": {\"field\": \"metric\", \"sort\": [\"Avg Voice Score (1-10) [Pearson]\", \"Weighted Ranking Score [Spearman]\"], \"type\": \"nominal\"}}}, {\"data\": {\"name\": \"data-59f2b1b765dab085e4ba1e557bab9669\"}, \"mark\": {\"type\": \"text\", \"color\": \"white\", \"fontSize\": 11, \"fontWeight\": \"normal\"}, \"encoding\": {\"text\": {\"field\": \"correlation\", \"format\": \".2f\", \"type\": \"quantitative\"}, \"x\": {\"field\": \"attribute\", \"sort\": [\"Age\", \"Mean pitch\", \"Standard deviation pitch\", \"Minimum pitch\", \"Maximum pitch\", \"Number of voice breaks\", \"Jitter (local)\", \"Shimmer (local)\", \"Mean harmonics-to-noise ratio dB\", \"H1-minus-H2\", \"MinI(dB)\", \"MaxI(dB)\", \"MeanI(dB)\", \"SDI(dB)\", \"MeanAbsSlope(dB/s)\", \"dur (s)\", \"phonationtime (s)\", \"speechrate (nsyll/dur)\", \"articulation rate (nsyll / phonationtime)\", \"speech rate words per minute\"], \"type\": \"nominal\"}, \"y\": {\"field\": \"metric\", \"sort\": [\"Avg Voice Score (1-10) [Pearson]\", \"Weighted Ranking Score [Spearman]\"], \"type\": \"nominal\"}}}], \"height\": 160, \"title\": {\"text\": [\"Speech Characteristics vs Survey Metrics\", \"Female Voices Only (Pearson / Spearman)\"], \"subtitle\": [\"Sample size: 455\"], \"subtitleColor\": \"gray\", \"subtitleFontSize\": 10, \"anchor\": \"start\"}, \"width\": 1100, \"$schema\": \"https://vega.github.io/schema/vega-lite/v6.1.0.json\", \"datasets\": {\"data-ff3d93a394a88d44a18ee07d1bece27c\": [{\"attribute\": \"Age\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.023565559711957996}, {\"attribute\": \"Age\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.23190841426097938}, {\"attribute\": \"Mean pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.19016169196951113}, {\"attribute\": \"Mean pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.19820624179302296}, {\"attribute\": \"Standard deviation pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.33721050044841366}, {\"attribute\": \"Standard deviation pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.12613124477737825}, {\"attribute\": \"Minimum pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.6155078897604899}, {\"attribute\": \"Minimum pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.21622499104693416}, {\"attribute\": \"Maximum pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.17733809220861338}, {\"attribute\": \"Maximum pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.6126374746329801}, {\"attribute\": \"Number of voice breaks\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.6316973759292629}, {\"attribute\": \"Number of voice breaks\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.027524094128159014}, {\"attribute\": \"Jitter (local)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.13206426212662656}, {\"attribute\": \"Jitter (local)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.05405624776173354}, {\"attribute\": \"Shimmer (local)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.3537250698256}, {\"attribute\": \"Shimmer (local)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.21622499104693416}, {\"attribute\": \"Mean harmonics-to-noise ratio dB\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.2539964392617356}, {\"attribute\": \"Mean harmonics-to-noise ratio dB\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.018018749253911177}, {\"attribute\": \"H1-minus-H2\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.5810981615794052}, {\"attribute\": \"H1-minus-H2\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.3964124835860459}, {\"attribute\": \"MinI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.19824072563494607}, {\"attribute\": \"MinI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.41443123283995714}, {\"attribute\": \"MaxI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.11491430922360052}, {\"attribute\": \"MaxI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.5585812268712466}, {\"attribute\": \"MeanI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.6893083031964921}, {\"attribute\": \"MeanI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.6363636363636364}, {\"attribute\": \"SDI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.6585847686635248}, {\"attribute\": \"SDI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.12613124477737825}, {\"attribute\": \"MeanAbsSlope(dB/s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.4478442793358411}, {\"attribute\": \"MeanAbsSlope(dB/s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.3603749850782236}, {\"attribute\": \"dur (s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.03526726467129442}, {\"attribute\": \"dur (s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.0900937462695559}, {\"attribute\": \"phonationtime (s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.029359105916164314}, {\"attribute\": \"phonationtime (s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.1801874925391118}, {\"attribute\": \"speechrate (nsyll/dur)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.0900937462695559}, {\"attribute\": \"articulation rate (nsyll / phonationtime)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.027420506142136723}, {\"attribute\": \"articulation rate (nsyll / phonationtime)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.1801874925391118}, {\"attribute\": \"speech rate words per minute\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.031236055978832018}, {\"attribute\": \"speech rate words per minute\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.0900937462695559}], \"data-46f3f309a17840278531f68da6ea94e1\": [{\"attribute\": \"Age\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.023565559711957996}, {\"attribute\": \"Age\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.23190841426097938}, {\"attribute\": \"Mean pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.19016169196951113}, {\"attribute\": \"Mean pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.19820624179302296}, {\"attribute\": \"Standard deviation pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.33721050044841366}, {\"attribute\": \"Standard deviation pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.12613124477737825}, {\"attribute\": \"Minimum pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.21622499104693416}, {\"attribute\": \"Maximum pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.17733809220861338}, {\"attribute\": \"Number of voice breaks\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.027524094128159014}, {\"attribute\": \"Jitter (local)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.13206426212662656}, {\"attribute\": \"Jitter (local)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.05405624776173354}, {\"attribute\": \"Shimmer (local)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.3537250698256}, {\"attribute\": \"Shimmer (local)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.21622499104693416}, {\"attribute\": \"Mean harmonics-to-noise ratio dB\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.2539964392617356}, {\"attribute\": \"Mean harmonics-to-noise ratio dB\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.018018749253911177}, {\"attribute\": \"H1-minus-H2\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.3964124835860459}, {\"attribute\": \"MinI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.19824072563494607}, {\"attribute\": \"MinI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.41443123283995714}, {\"attribute\": \"MaxI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.11491430922360052}, {\"attribute\": \"SDI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.12613124477737825}, {\"attribute\": \"MeanAbsSlope(dB/s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.4478442793358411}, {\"attribute\": \"MeanAbsSlope(dB/s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.3603749850782236}, {\"attribute\": \"dur (s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.03526726467129442}, {\"attribute\": \"dur (s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.0900937462695559}, {\"attribute\": \"phonationtime (s)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.029359105916164314}, {\"attribute\": \"phonationtime (s)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.1801874925391118}, {\"attribute\": \"speechrate (nsyll/dur)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.0900937462695559}, {\"attribute\": \"articulation rate (nsyll / phonationtime)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.027420506142136723}, {\"attribute\": \"articulation rate (nsyll / phonationtime)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.1801874925391118}, {\"attribute\": \"speech rate words per minute\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.031236055978832018}, {\"attribute\": \"speech rate words per minute\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.0900937462695559}], \"data-59f2b1b765dab085e4ba1e557bab9669\": [{\"attribute\": \"Minimum pitch\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.6155078897604899}, {\"attribute\": \"Maximum pitch\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": 0.6126374746329801}, {\"attribute\": \"Number of voice breaks\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": 0.6316973759292629}, {\"attribute\": \"H1-minus-H2\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.5810981615794052}, {\"attribute\": \"MaxI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.5585812268712466}, {\"attribute\": \"MeanI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.6893083031964921}, {\"attribute\": \"MeanI(dB)\", \"metric\": \"Weighted Ranking Score [Spearman]\", \"correlation\": -0.6363636363636364}, {\"attribute\": \"SDI(dB)\", \"metric\": \"Avg Voice Score (1-10) [Pearson]\", \"correlation\": -0.6585847686635248}]}}, {\"mode\": \"vega-lite\"});\n",
|
|
"</script>"
|
|
],
|
|
"text/plain": [
|
|
"alt.LayerChart(...)"
|
|
]
|
|
},
|
|
"execution_count": 45,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"# --- Plot: Female voices ---\n",
|
|
"S.plot_speech_attribute_correlation(\n",
|
|
" corr_female,\n",
|
|
" title=\"Speech Characteristics vs Survey Metrics<br>Female Voices Only (Pearson / Spearman)\",\n",
|
|
" filename=\"speech_attr_vs_survey_correlation_female\",\n",
|
|
")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 47,
|
|
"id": "8e2fbc25",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"=== MALE | Age vs Avg Voice Score (1-10) [Pearson] | r = 0.793 ===\n",
|
|
"shape: (9, 3)\n",
|
|
"┌───────┬─────┬────────────────────────┐\n",
|
|
"│ Voice ┆ Age ┆ Avg Voice Score (1-10) │\n",
|
|
"│ --- ┆ --- ┆ --- │\n",
|
|
"│ str ┆ i64 ┆ f64 │\n",
|
|
"╞═══════╪═════╪════════════════════════╡\n",
|
|
"│ V81 ┆ 28 ┆ 6.683007 │\n",
|
|
"│ V54 ┆ 36 ┆ 6.67734 │\n",
|
|
"│ V16 ┆ 40 ┆ 6.921053 │\n",
|
|
"│ V74 ┆ 41 ┆ 6.89196 │\n",
|
|
"│ V34 ┆ 42 ┆ 7.153005 │\n",
|
|
"│ V88 ┆ 42 ┆ 6.916667 │\n",
|
|
"│ V69 ┆ 43 ┆ 7.219577 │\n",
|
|
"│ V45 ┆ 43 ┆ 7.062189 │\n",
|
|
"│ V86 ┆ 62 ┆ 7.264444 │\n",
|
|
"└───────┴─────┴────────────────────────┘\n",
|
|
"\n",
|
|
"=== MALE | Shimmer (local) vs Avg Voice Score (1-10) [Pearson] | r = 0.813 ===\n",
|
|
"shape: (9, 3)\n",
|
|
"┌───────┬─────────────────┬────────────────────────┐\n",
|
|
"│ Voice ┆ Shimmer (local) ┆ Avg Voice Score (1-10) │\n",
|
|
"│ --- ┆ --- ┆ --- │\n",
|
|
"│ str ┆ f64 ┆ f64 │\n",
|
|
"╞═══════╪═════════════════╪════════════════════════╡\n",
|
|
"│ V88 ┆ 9.07 ┆ 6.916667 │\n",
|
|
"│ V74 ┆ 9.09 ┆ 6.89196 │\n",
|
|
"│ V54 ┆ 9.11 ┆ 6.67734 │\n",
|
|
"│ V81 ┆ 9.13 ┆ 6.683007 │\n",
|
|
"│ V16 ┆ 9.38 ┆ 6.921053 │\n",
|
|
"│ V34 ┆ 10.07 ┆ 7.153005 │\n",
|
|
"│ V86 ┆ 10.28 ┆ 7.264444 │\n",
|
|
"│ V45 ┆ 10.39 ┆ 7.062189 │\n",
|
|
"│ V69 ┆ 11.52 ┆ 7.219577 │\n",
|
|
"└───────┴─────────────────┴────────────────────────┘\n",
|
|
"\n",
|
|
"=== MALE | Mean harmonics-to-noise ratio dB vs Avg Voice Score (1-10) [Pearson] | r = -0.738 ===\n",
|
|
"shape: (9, 3)\n",
|
|
"┌───────┬─────────────────────────────────┬────────────────────────┐\n",
|
|
"│ Voice ┆ Mean harmonics-to-noise ratio … ┆ Avg Voice Score (1-10) │\n",
|
|
"│ --- ┆ --- ┆ --- │\n",
|
|
"│ str ┆ f64 ┆ f64 │\n",
|
|
"╞═══════╪═════════════════════════════════╪════════════════════════╡\n",
|
|
"│ V16 ┆ 7.544 ┆ 6.921053 │\n",
|
|
"│ V34 ┆ 7.671 ┆ 7.153005 │\n",
|
|
"│ V69 ┆ 7.763 ┆ 7.219577 │\n",
|
|
"│ V86 ┆ 8.092 ┆ 7.264444 │\n",
|
|
"│ V45 ┆ 8.647 ┆ 7.062189 │\n",
|
|
"│ V74 ┆ 8.732 ┆ 6.89196 │\n",
|
|
"│ V88 ┆ 9.009 ┆ 6.916667 │\n",
|
|
"│ V81 ┆ 9.026 ┆ 6.683007 │\n",
|
|
"│ V54 ┆ 9.598 ┆ 6.67734 │\n",
|
|
"└───────┴─────────────────────────────────┴────────────────────────┘\n",
|
|
"\n",
|
|
"=== MALE | Mean harmonics-to-noise ratio dB vs Weighted Ranking Score [Spearman] | r = -0.767 ===\n",
|
|
"shape: (10, 3)\n",
|
|
"┌───────┬─────────────────────────────────┬────────────────┐\n",
|
|
"│ Voice ┆ Mean harmonics-to-noise ratio … ┆ Weighted Score │\n",
|
|
"│ --- ┆ --- ┆ --- │\n",
|
|
"│ str ┆ f64 ┆ i64 │\n",
|
|
"╞═══════╪═════════════════════════════════╪════════════════╡\n",
|
|
"│ V16 ┆ 7.544 ┆ 156 │\n",
|
|
"│ V34 ┆ 7.671 ┆ 128 │\n",
|
|
"│ V69 ┆ 7.763 ┆ 121 │\n",
|
|
"│ V86 ┆ 8.092 ┆ 76 │\n",
|
|
"│ V46 ┆ 8.625 ┆ 148 │\n",
|
|
"│ V45 ┆ 8.647 ┆ 118 │\n",
|
|
"│ V74 ┆ 8.732 ┆ 117 │\n",
|
|
"│ V88 ┆ 9.009 ┆ 97 │\n",
|
|
"│ V81 ┆ 9.026 ┆ 69 │\n",
|
|
"│ V54 ┆ 9.598 ┆ 116 │\n",
|
|
"└───────┴─────────────────────────────────┴────────────────┘\n",
|
|
"\n",
|
|
"=== FEMALE: no correlations with |r| > 0.7 ===\n",
|
|
"\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"# Inspect underlying data for all |r| > 0.7 correlations in gender subgroups\n",
|
|
"THRESHOLD = 0.7\n",
|
|
"\n",
|
|
"metric_col_map = {\n",
|
|
" \"Avg Voice Score (1-10) [Pearson]\": \"Avg Voice Score (1-10)\",\n",
|
|
" \"Weighted Ranking Score [Spearman]\": \"Weighted Score\",\n",
|
|
"}\n",
|
|
"\n",
|
|
"for gender, corr_df in [(\"MALE\", corr_male), (\"FEMALE\", corr_female)]:\n",
|
|
" strong = corr_df.filter(pl.col(\"correlation\").abs() > THRESHOLD)\n",
|
|
" if strong.height == 0:\n",
|
|
" print(f\"=== {gender}: no correlations with |r| > {THRESHOLD} ===\\n\")\n",
|
|
" continue\n",
|
|
"\n",
|
|
" subset = merged.filter(pl.col(\"Gender\") == gender.lower())\n",
|
|
"\n",
|
|
" for row in strong.iter_rows(named=True):\n",
|
|
" attr = row[\"attribute\"]\n",
|
|
" metric_label = row[\"metric\"]\n",
|
|
" r = row[\"correlation\"]\n",
|
|
" survey_col = metric_col_map[metric_label]\n",
|
|
"\n",
|
|
" table = (\n",
|
|
" subset\n",
|
|
" .select([\"Voice\", attr, survey_col])\n",
|
|
" .drop_nulls()\n",
|
|
" .sort(attr)\n",
|
|
" )\n",
|
|
" print(f\"=== {gender} | {attr} vs {metric_label} | r = {r:.3f} ===\")\n",
|
|
" print(table)\n",
|
|
" print()"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 48,
|
|
"id": "959945f2",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Saved 4 sheets to figures/2-4-26/strong_correlations_by_gender.xlsx\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"import xlsxwriter\n",
|
|
"from pathlib import Path\n",
|
|
"\n",
|
|
"out_path = Path(S.fig_save_dir) / \"strong_correlations_by_gender.xlsx\"\n",
|
|
"\n",
|
|
"with xlsxwriter.Workbook(str(out_path)) as wb:\n",
|
|
" sheet_count = 0\n",
|
|
" for gender, corr_df in [(\"Male\", corr_male), (\"Female\", corr_female)]:\n",
|
|
" strong = corr_df.filter(pl.col(\"correlation\").abs() > THRESHOLD)\n",
|
|
" if strong.height == 0:\n",
|
|
" continue\n",
|
|
"\n",
|
|
" subset = merged.filter(pl.col(\"Gender\") == gender.lower())\n",
|
|
"\n",
|
|
" for row in strong.iter_rows(named=True):\n",
|
|
" attr = row[\"attribute\"]\n",
|
|
" metric_label = row[\"metric\"]\n",
|
|
" r = row[\"correlation\"]\n",
|
|
" survey_col = metric_col_map[metric_label]\n",
|
|
"\n",
|
|
" table = subset.select([\"Voice\", attr, survey_col]).drop_nulls().sort(attr)\n",
|
|
"\n",
|
|
" # Sheet name: max 31 chars for Excel\n",
|
|
" short_metric = \"Score\" if \"Pearson\" in metric_label else \"Rank\"\n",
|
|
" sheet_name = f\"{gender}_{attr[:18]}_{short_metric}\"[:31]\n",
|
|
"\n",
|
|
" ws = wb.add_worksheet(sheet_name)\n",
|
|
" # Header row with context\n",
|
|
" bold = wb.add_format({\"bold\": True})\n",
|
|
" ws.write(0, 0, f\"{gender} | {attr} vs {metric_label} | r = {r:.3f}\", bold)\n",
|
|
"\n",
|
|
" # Column headers\n",
|
|
" for ci, col_name in enumerate(table.columns):\n",
|
|
" ws.write(2, ci, col_name, bold)\n",
|
|
"\n",
|
|
" # Data rows\n",
|
|
" for ri, data_row in enumerate(table.iter_rows()):\n",
|
|
" for ci, val in enumerate(data_row):\n",
|
|
" ws.write(3 + ri, ci, val)\n",
|
|
"\n",
|
|
" # Auto-fit column widths\n",
|
|
" for ci, col_name in enumerate(table.columns):\n",
|
|
" max_len = max(len(str(col_name)), *(len(str(v)) for v in table.get_column(col_name).to_list()))\n",
|
|
" ws.set_column(ci, ci, max_len + 2)\n",
|
|
"\n",
|
|
" sheet_count += 1\n",
|
|
"\n",
|
|
"print(f\"Saved {sheet_count} sheets to {out_path}\")"
|
|
]
|
|
}
|
|
],
|
|
"metadata": {
|
|
"kernelspec": {
|
|
"display_name": "phase-3-quant",
|
|
"language": "python",
|
|
"name": "python3"
|
|
},
|
|
"language_info": {
|
|
"codemirror_mode": {
|
|
"name": "ipython",
|
|
"version": 3
|
|
},
|
|
"file_extension": ".py",
|
|
"mimetype": "text/x-python",
|
|
"name": "python",
|
|
"nbconvert_exporter": "python",
|
|
"pygments_lexer": "ipython3",
|
|
"version": "3.12.10"
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 5
|
|
}
|