Looping through a set of wav files using Whisper

[rivkahcarl]

Hello Whisper community, Happy new year!

I was wondering if someone could help me with a bit of python and Whisper.
We have created a script to loop through a folder of wav files:

import whisper
import os

# Specify the path to the directory containing your files
folder_path = '/Users/mcarl/Desktop/whisper_files'
model = whisper.load_model("medium")

# Iterate through each file in the directory
for filename in os.listdir(folder_path):
    # Print or perform any other operation with the file name
    print(filename)
    print(folder_path+"/"+filename)
    result = model.transcribe(folder_path+"/"+filename)
    print(result["text"])

but for some reason am getting the following error.

Traceback (most recent call last):
  File "/Users/mcarl/Desktop/whisper_script.py", line 17, in <module>
    result = model.transcribe(folder_path)
  File "/Users/mcarl/.pyenv/versions/3.9.2/lib/python3.9/site-packages/whisper/transcribe.py", line 133, in transcribe
    mel = log_mel_spectrogram(audio, model.dims.n_mels, padding=N_SAMPLES)
  File "/Users/mcarl/.pyenv/versions/3.9.2/lib/python3.9/site-packages/whisper/audio.py", line 140, in log_mel_spectrogram
    audio = load_audio(audio)
  File "/Users/mcarl/.pyenv/versions/3.9.2/lib/python3.9/site-packages/whisper/audio.py", line 60, in load_audio
    raise RuntimeError(f"Failed to load audio: {e.stderr.decode()}") from e
RuntimeError: Failed to load audio: ffmpeg version 6.1.1 Copyright (c) 2000-2023 the FFmpeg developers
  built with Apple clang version 15.0.0 (clang-1500.1.0.2.5)
  configuration: --prefix=/opt/homebrew/Cellar/ffmpeg/6.1.1_1 --enable-shared --enable-pthreads --enable-version3 --cc=clang --host-cflags= --host-ldflags='-Wl,-ld_classic' --enable-ffplay --enable-gnutls --enable-gpl --enable-libaom --enable-libaribb24 --enable-libbluray --enable-libdav1d --enable-libharfbuzz --enable-libjxl --enable-libmp3lame --enable-libopus --enable-librav1e --enable-librist --enable-librubberband --enable-libsnappy --enable-libsrt --enable-libsvtav1 --enable-libtesseract --enable-libtheora --enable-libvidstab --enable-libvmaf --enable-libvorbis --enable-libvpx --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxml2 --enable-libxvid --enable-lzma --enable-libfontconfig --enable-libfreetype --enable-frei0r --enable-libass --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-libspeex --enable-libsoxr --enable-libzmq --enable-libzimg --disable-libjack --disable-indev=jack --enable-videotoolbox --enable-audiotoolbox --enable-neon
  libavutil      58. 29.100 / 58. 29.100
  libavcodec     60. 31.102 / 60. 31.102
  libavformat    60. 16.100 / 60. 16.100
  libavdevice    60.  3.100 / 60.  3.100
  libavfilter     9. 12.100 /  9. 12.100
  libswscale      7.  5.100 /  7.  5.100
  libswresample   4. 12.100 /  4. 12.100
  libpostproc    57.  3.100 / 57.  3.100
[in#0 @ 0x600001edc000] Error opening input: Is a directory

Error opening input file /Users/mcarl/Desktop/whisper_files.
Error opening input files: Is a directory

I think the solution is likely something simple related to how the files get loaded and was hoping someone could help.

thank you in advance!

[Purfview]

Try:

result = model.transcribe(folder_path +os.path.sep+ filename)

[rivkahcarl]

thanks for the response,
Unfortunately that seems to be pulling .DS_Store files.
What I am struggling to understand is this an issue with just looping through string names of the files, or does the .transcribe function somehow need to open the file which is being prevented by .os.

Thats where I am a tad confused.

/Users/mcarl/Desktop/whisper_files/.DS_Store

/Users/mcarl/.pyenv/versions/3.9.2/lib/python3.9/site-packages/whisper/transcribe.py:126: UserWarning: FP16 is not supported on CPU; using FP32 instead

  warnings.warn("FP16 is not supported on CPU; using FP32 instead")

Traceback (most recent call last):

  File "/Users/mcarl/.pyenv/versions/3.9.2/lib/python3.9/site-packages/whisper/audio.py", line 58, in load_audio

    out = run(cmd, capture_output=True, check=True).stdout

  File "/Users/mcarl/.pyenv/versions/3.9.2/lib/python3.9/subprocess.py", line 528, in run

    raise CalledProcessError(retcode, process.args,

subprocess.CalledProcessError: Command '['ffmpeg', '-nostdin', '-threads', '0', '-i', '/Users/mcarl/Desktop/whisper_files/.DS_Store', '-f', 's16le', '-ac', '1', '-acodec', 'pcm_s16le', '-ar', '16000', '-']' returned non-zero exit status 183.

The above exception was the direct cause of the following exception:
Traceback (most recent call last):

  File "/Users/mcarl/Desktop/whisper_script.py", line 17, in <module>

    result = model.transcribe(folder_path +os.path.sep+ filename)

  File "/Users/mcarl/.pyenv/versions/3.9.2/lib/python3.9/site-packages/whisper/transcribe.py", line 133, in transcribe

    mel = log_mel_spectrogram(audio, model.dims.n_mels, padding=N_SAMPLES)

  File "/Users/mcarl/.pyenv/versions/3.9.2/lib/python3.9/site-packages/whisper/audio.py", line 140, in log_mel_spectrogram

    audio = load_audio(audio)

  File "/Users/mcarl/.pyenv/versions/3.9.2/lib/python3.9/site-packages/whisper/audio.py", line 60, in load_audio

    raise RuntimeError(f"Failed to load audio: {e.stderr.decode()}") from e

RuntimeError: Failed to load audio: ffmpeg version 6.1.1 Copyright (c) 2000-2023 the FFmpeg developers

  built with Apple clang version 15.0.0 (clang-1500.1.0.2.5)

  configuration: --prefix=/opt/homebrew/Cellar/ffmpeg/6.1.1_1 --enable-shared --enable-pthreads --enable-version3 --cc=clang --host-cflags= --host-ldflags='-Wl,-ld_classic' --enable-ffplay --enable-gnutls --enable-gpl --enable-libaom --enable-libaribb24 --enable-libbluray --enable-libdav1d --enable-libharfbuzz --enable-libjxl --enable-libmp3lame --enable-libopus --enable-librav1e --enable-librist --enable-librubberband --enable-libsnappy --enable-libsrt --enable-libsvtav1 --enable-libtesseract --enable-libtheora --enable-libvidstab --enable-libvmaf --enable-libvorbis --enable-libvpx --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxml2 --enable-libxvid --enable-lzma --enable-libfontconfig --enable-libfreetype --enable-frei0r --enable-libass --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-libspeex --enable-libsoxr --enable-libzmq --enable-libzimg --disable-libjack --disable-indev=jack --enable-videotoolbox --enable-audiotoolbox --enable-neon

  libavutil      58. 29.100 / 58. 29.100

  libavcodec     60. 31.102 / 60. 31.102
  libavformat    60. 16.100 / 60. 16.100
  libavdevice    60.  3.100 / 60.  3.100
  libavfilter     9. 12.100 /  9. 12.100
  libswscale      7.  5.100 /  7.  5.100
  libswresample   4. 12.100 /  4. 12.100
  libpostproc    57.  3.100 / 57.  3.100
[in#0 @ 0x600000ec4000] Error opening input: Invalid data found when processing input
Error opening input file /Users/mcarl/Desktop/whisper_files/.DS_Store.
Error opening input files: Invalid data found when processing input

[glangford]

Note that os.listdir will give directories as well as files.

[Purfview]

Try:

media_ext = ['.3gp', '.aac', '.ac3', '.aif', '.aifc', '.aiff', '.amr', '.ape', '.asf', '.asx', '.au', '.avi', '.bdmv', '.bwf', '.caf', '.dat', '.dts', '.dtshd', '.eac3', '.eb3', '.ec3', '.f4v', '.flac', '.fli', '.flv', '.l16', '.m1a', '.m2a', '.m2ts', '.m4a', '.m4v', '.mid', '.mka', '.mkv', '.mlp', '.mod', '.mov', '.mp1', '.mp2', '.mp3', '.mp4', '.mpa', '.mpc', '.mpeg', '.mpg', '.ofr', '.oga', '.ogg', '.opus', '.opus', '.pcm', '.qt', '.ra', '.ram', '.rm', '.snd', '.spx', '.stm', '.tak', '.thd', '.ts', '.tta', '.vob', '.voc', '.vqf', '.wav', '.wave', '.webm', '.wma', '.wmv', '.wv']
folder_path = '/Users/mcarl/Desktop/whisper_files'

if os.path.isdir(folder_path):
    for root, dirs, files in os.walk(folder_path):
        for filename in files:
            if os.path.splitext(filename)[1].lower() in media_ext:
                filepath = root + os.sep + filename
                result = model.transcribe(filepath)

[rivkahcarl]

Thank you so much!
This worked correctly and output the result text.
Was the issue that we had the wrong root for full access to the file?

[Purfview]

Dunno about macOS stuff, the problem from your last paste was that "DS_Store" is not a media file.
In my example it looks only for the files with extensions specified in media_ext list.

Error opening input file /Users/mcarl/Desktop/whisper_files/.DS_Store