Doesn't work properly on win10

[fcsnsu]

PS D:\Download\Secrit> whisper csct-013.mp3 --model medium
100%|█████████████████████████████████████| 1.42G/1.42G [02:15<00:00, 11.3MiB/s]
D:\Date\python\Lib\site-packages\whisper\transcribe.py:114: 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 "D:\Date\python\Lib\site-packages\whisper\audio.py", line 59, in load_audio
    out = run(cmd, capture_output=True, check=True).stdout
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\Date\python\Lib\subprocess.py", line 571, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ffmpeg', '-nostdin', '-threads', '0', '-i', 'csct-013.mp3', '-f', 's16le', '-ac', '1', '-acodec', 'pcm_s16le', '-ar', '16000', '-']' returned non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "<frozen runpy>", line 198, in _run_module_as_main
  File "<frozen runpy>", line 88, in _run_code
  File "D:\Date\python\Scripts\whisper.exe\__main__.py", line 7, in <module>
  File "D:\Date\python\Lib\site-packages\whisper\transcribe.py", line 456, in cli
    result = transcribe(model, audio_path, temperature=temperature, **args)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\Date\python\Lib\site-packages\whisper\transcribe.py", line 121, in transcribe
    mel = log_mel_spectrogram(audio, padding=N_SAMPLES)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\Date\python\Lib\site-packages\whisper\audio.py", line 140, in log_mel_spectrogram
    audio = load_audio(audio)
            ^^^^^^^^^^^^^^^^^
  File "D:\Date\python\Lib\site-packages\whisper\audio.py", line 61, in load_audio
    raise RuntimeError(f"Failed to load audio: {e.stderr.decode()}") from e
RuntimeError: Failed to load audio: ffmpeg version N-94813-g85386c36e3-ffmpeg-for-N_m3u8DL-CLI Copyright (c) 2000-2019 the FFmpeg developers
  built with gcc 8.2.0 (GCC)
  libavutil      56. 35.100 / 56. 35.100
  libavcodec     58. 56.101 / 58. 56.101
  libavformat    58. 32.104 / 58. 32.104
  libavfilter     7. 58.102 /  7. 58.102
  libavresample   4.  0.  0 /  4.  0.  0
  libswresample   3.  6.100 /  3.  6.100
Input #0, mp3, from 'csct-013.mp3':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf57.56.100
  Duration: 02:00:51.48, start: 0.023021, bitrate: 112 kb/s
    Stream #0:0: Audio: mp3, 48000 Hz, stereo, fltp, 112 kb/s
    Metadata:
      encoder         : Lavc57.64
[NULL @ 02ac87c0] Requested output format 's16le' is not a suitable output format
pipe:: Invalid argument

[glangford]

Whisper requires ffmpeg to process the audio. Try ffmpeg --version from the command line. If that does not work, ensure that ffmpeg is installed (search other discussions, this is a common problem).

[fcsnsu]

I have ffmpeg installed.

PS C:\Users\FCSN> ffmpeg --version
ffmpeg version N-94813-g85386c36e3-ffmpeg-for-N_m3u8DL-CLI Copyright (c) 2000-2019 the FFmpeg developers
  built with gcc 8.2.0 (GCC)
  libavutil      56. 35.100 / 56. 35.100
  libavcodec     58. 56.101 / 58. 56.101
  libavformat    58. 32.104 / 58. 32.104
  libavfilter     7. 58.102 /  7. 58.102
  libavresample   4.  0.  0 /  4.  0.  0
  libswresample   3.  6.100 /  3.  6.100
Unrecognized option '-version'.
Error splitting the argument list: Option not found

[glangford]

Can you get it to work with a different audio file

[fcsnsu]

no,it still doesn't work

D:\Date\python\Lib\site-packages\whisper\transcribe.py:114: 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 "D:\Date\python\Lib\site-packages\whisper\audio.py", line 59, in load_audio
    out = run(cmd, capture_output=True, check=True).stdout
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\Date\python\Lib\subprocess.py", line 571, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ffmpeg', '-nostdin', '-threads', '0', '-i', 'test.mp3', '-f', 's16le', '-ac', '1', '-acodec', 'pcm_s16le', '-ar', '16000', '-']' returned non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "<frozen runpy>", line 198, in _run_module_as_main
  File "<frozen runpy>", line 88, in _run_code
  File "D:\Date\python\Scripts\whisper.exe\__main__.py", line 7, in <module>
  File "D:\Date\python\Lib\site-packages\whisper\transcribe.py", line 456, in cli
    result = transcribe(model, audio_path, temperature=temperature, **args)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\Date\python\Lib\site-packages\whisper\transcribe.py", line 121, in transcribe
    mel = log_mel_spectrogram(audio, padding=N_SAMPLES)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\Date\python\Lib\site-packages\whisper\audio.py", line 140, in log_mel_spectrogram
    audio = load_audio(audio)
            ^^^^^^^^^^^^^^^^^
  File "D:\Date\python\Lib\site-packages\whisper\audio.py", line 61, in load_audio
    raise RuntimeError(f"Failed to load audio: {e.stderr.decode()}") from e
RuntimeError: Failed to load audio: ffmpeg version N-94813-g85386c36e3-ffmpeg-for-N_m3u8DL-CLI Copyright (c) 2000-2019 the FFmpeg developers
  built with gcc 8.2.0 (GCC)
  libavutil      56. 35.100 / 56. 35.100
  libavcodec     58. 56.101 / 58. 56.101
  libavformat    58. 32.104 / 58. 32.104
  libavfilter     7. 58.102 /  7. 58.102
  libavresample   4.  0.  0 /  4.  0.  0
  libswresample   3.  6.100 /  3.  6.100
Input #0, mp3, from 'test.mp3':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    comment         : Unknown
    encoder         : Lavf57.56.100
  Duration: 00:20:50.45, start: 0.023021, bitrate: 165 kb/s
    Stream #0:0: Audio: mp3, 48000 Hz, stereo, fltp, 165 kb/s
    Metadata:
      encoder         : Lavc57.64
[NULL @ 009fe840] Requested output format 's16le' is not a suitable output format
pipe:: Invalid argument

[glangford]

Requested output format 's16le' is not a suitable output format

Is an error triggered by ffmpeg (haven't seen that before myself, and I am not a Win 10 user).

[fcsnsu]

I tried reinstalling ffmpeg and it's back up and working fast now, thanks a lot!