voice to text

transcribe.py

import os
import sys
import glob
import whisper

def find_audio_files(directory):
    # mp3 and m4a, case-insensitive
    patterns = ["*.mp3", "*.MP3", "*.m4a", "*.M4A"]
    files = []
    for pattern in patterns:
        files.extend(glob.glob(os.path.join(directory, pattern)))
    return sorted(files)

def main():
    if len(sys.argv) < 2:
        print("Usage: python batch_transcribe.py /path/to/dir [model_name]")
        sys.exit(1)

    audio_dir = sys.argv[1]
    model_name = sys.argv[2] if len(sys.argv) >= 3 else "small"  # you can change to "medium"

    if not os.path.isdir(audio_dir):
        print(f"Error: '{audio_dir}' is not a directory.")
        sys.exit(1)

    print(f"Loading Whisper model: {model_name}")
    model = whisper.load_model(model_name)

    audio_files = find_audio_files(audio_dir)
    if not audio_files:
        print("No .mp3 or .m4a files found in that directory.")
        sys.exit(0)

    for path in audio_files:
        base, _ = os.path.splitext(path)
        txt_path = base + ".txt"

        if os.path.exists(txt_path):
            print(f"Skipping (txt already exists): {os.path.basename(path)}")
            continue

        print(f"Transcribing: {os.path.basename(path)}")
        try:
            result = model.transcribe(path)
            text = result.get("text", "").strip()
            with open(txt_path, "w", encoding="utf-8") as f:
                f.write(text + "\n")
            print(f"Saved: {os.path.basename(txt_path)}")
        except Exception as e:
            print(f"Failed on {os.path.basename(path)}: {e}")

if __name__ == "__main__":
    main()
 



# python batch_transcribe.py /path/to/your/audio medium