This commit is contained in:
Hoang Dinh 2024-09-13 15:55:12 +03:00 committed by GitHub
commit 63cfdf1431
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 3 additions and 1 deletions

View file

@ -143,7 +143,7 @@ def encode_audio_stream(path, ffmpeg_path, audio_stream_language=None):
logger.debug(f"Whisper will only use the {audio_stream_language} audio stream for {path}")
inp = inp[f'a:m:language:{audio_stream_language}']
out, _ = inp.output("-", format="s16le", acodec="pcm_s16le", ac=1, ar=16000) \
out, _ = inp.output("-", format="s16le", acodec="pcm_s16le", ac=1, ar=16000, af="aresample=async=1") \
.run(cmd=[ffmpeg_path, "-nostdin"], capture_stdout=True, capture_stderr=True)
except ffmpeg.Error as e:

View file

@ -367,6 +367,8 @@ class VideoSpeechTransformer(TransformerMixin):
"1",
"-acodec",
"pcm_s16le",
"-af",
"aresample=async=1",
"-ar",
str(self.frame_rate),
"-",