speech2text.py

import os, re

import whisper
from whisper.utils import get_writer

model=whisper.load_model("base").to('cpu')

def get_file_stem(file_name):
    # Use a regular expression to match and extract the file stem
    match = re.match(r'^(.*/)?([^.]+)\.\w+$', file_name)
    if match:
        return match.group(2)
    else:
        return None
    
def transcribeAndSave(file,outputPath):
    result = model.transcribe(file)
    writer = get_writer('tsv',outputPath)
    writer(result,get_file_stem(file),None)

def transcribeAndSaveFolder(fileFolder,outputpath):
    os.makedirs(outputpath, exist_ok=True)
    for file in os.listdir(fileFolder):
        filepath=os.path.join(fileFolder,file)
        transcribeAndSave(filepath,outputpath)