Spaces:
Running
Running
File size: 4,125 Bytes
547836e 8e62829 547836e |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 |
from utils.load_csv import download_csv
def box_plot_data(ASR_model):
# Load the CSV file
csv_result = f'test_with_{ASR_model.replace("/","_")}_WER.csv'
df = download_csv(csv_result)
# Display actual column names to check for issues
print(df.columns)
# Trim column names of any leading or trailing spaces
df.columns = df.columns.str.strip()
# Extract WER values for Male and Female
wer_Gender = {
"Male": df[df["gender"] == "male"]["WER"].tolist(),
"Female": df[df["gender"] == "female"]["WER"].tolist()
}
wer_SEG = {
"Low": df[df["socioeconomic_bkgd"] == "Low"]["WER"].tolist(),
"Affluent": df[df["socioeconomic_bkgd"] == "Affluent"]["WER"].tolist(),
"Medium": df[df["socioeconomic_bkgd"] == "Medium"]["WER"].tolist(),
}
wer_Ethnicity = {
"Asian, South Asian or Asian American": df[df["ethnicity"] == "Asian, South Asian or Asian American"]["WER"].tolist(),
"Black or African American": df[df["ethnicity"] == "Black or African American"]["WER"].tolist(),
"Hispanic, Latino, or Spanish": df[df["ethnicity"] == "Hispanic, Latino, or Spanish"]["WER"].tolist(),
"Middle Eastern or North African": df[df["ethnicity"] == "Middle Eastern or North African"]["WER"].tolist(),
"Native American, American Indian, or Alaska Native": df[df["ethnicity"] == "Native American, American Indian, or Alaska Native"]["WER"].tolist(),
"Native Hawaiian or Other Pacific Islander": df[df["ethnicity"] == "Native Hawaiian or Other Pacific Islander"]["WER"].tolist(),
"White": df[df["ethnicity"] == "White"]["WER"].tolist(),
}
wer_Language = {
"English": df[df["first_language"] == "English"]["WER"].tolist(),
"German": df[df["first_language"] == "German"]["WER"].tolist(),
"French": df[df["first_language"] == "French"]["WER"].tolist(),
"Arabic": df[df["first_language"] == "Arabic"]["WER"].tolist(),
"Cantonese": df[df["first_language"] == "Cantonese"]["WER"].tolist(),
"Creole": df[df["first_language"] == "Creole"]["WER"].tolist(),
"Dutch": df[df["first_language"] == "Dutch"]["WER"].tolist(),
"English/Turkish": df[df["first_language"] == "English/Turkish"]["WER"].tolist(),
"Filipino": df[df["first_language"] == "Filipino"]["WER"].tolist(),
"Hindi": df[df["first_language"] == "Hindi"]["WER"].tolist(),
"Hmong": df[df["first_language"] == "Hmong"]["WER"].tolist(),
"Hindi": df[df["first_language"] == "Hindi"]["WER"].tolist(),
"Indonesian": df[df["first_language"] == "Indonesian"]["WER"].tolist(),
"Italian": df[df["first_language"] == "Italian"]["WER"].tolist(),
"Japanese": df[df["first_language"] == "Japanese"]["WER"].tolist(),
"Korean": df[df["first_language"] == "Korean"]["WER"].tolist(),
"Laotian": df[df["first_language"] == "Laotian"]["WER"].tolist(),
"Malay": df[df["first_language"] == "Malay"]["WER"].tolist(),
"Malaysian": df[df["first_language"] == "Malaysian"]["WER"].tolist(),
"Mandarin": df[df["first_language"] == "Mandarin"]["WER"].tolist(),
"Marathi": df[df["first_language"] == "Marathi"]["WER"].tolist(),
"Nepali": df[df["first_language"] == "Nepali"]["WER"].tolist(),
"Other": df[df["first_language"] == "Other"]["WER"].tolist(),
"Portuguese": df[df["first_language"] == "Portuguese"]["WER"].tolist(),
"Russian": df[df["first_language"] == "Russian"]["WER"].tolist(),
"Spanish": df[df["first_language"] == "Spanish"]["WER"].tolist(),
"Tagalog": df[df["first_language"] == "Tagalog"]["WER"].tolist(),
"Turkish": df[df["first_language"] == "Turkish"]["WER"].tolist(),
"Russian": df[df["first_language"] == "Russian"]["WER"].tolist(),
"Ukrainian": df[df["first_language"] == "Ukrainian"]["WER"].tolist(),
"Urdu": df[df["first_language"] == "Urdu"]["WER"].tolist(),
"Vietnamese": df[df["first_language"] == "Vietnamese"]["WER"].tolist(),
}
return wer_Gender, wer_SEG, wer_Ethnicity, wer_Language |