extern/whisper.cpp
1
0
Fork 1
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-05-17 22:51:11 +02:00
Code Issues Packages Projects Releases Wiki Activity

examples : add FFmpeg v7.0 support to ffmpeg-transcode.cpp (#3038)

Browse Source 
FFmpeg introduced a new channel layout API that uses `AVChannelLayout`
interface in v6.0. It subsequently dropped the old bitmask-based API
in v7.0.

This updates decode_audio() to support the new channel layout API,
so that we can compile `whisper-cli` and `whisper-server` with FFmpeg
v7.0 or later.

Tested on on Ubuntu 24.10 with FFmpeg v7.0.2.

Signed-off-by: Fujimoto Seiji <fujimoto@ceptord.net>
This commit is contained in:
Fujimoto Seiji 2025-04-15 13:09:00 +09:00 committed by GitHub
parent 2a2d21c75d
commit f8a3509b6d
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 15 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
examples/ffmpeg-transcode.cpp
View File

@ -249,6 +249,20 @@ static int decode_audio(struct audio_buffer *audio_buf, s16 **data, int *size)
/* prepare resampler */ /* prepare resampler */
swr = swr_alloc(); swr = swr_alloc();
#if LIBAVCODEC_VERSION_MAJOR > 60
AVChannelLayout in_ch_layout = codec->ch_layout;
AVChannelLayout out_ch_layout = AV_CHANNEL_LAYOUT_MONO;
/* Set the source audio layout as-is */
av_opt_set_chlayout(swr, "in_chlayout", &in_ch_layout, 0);
av_opt_set_int(swr, "in_sample_rate", codec->sample_rate, 0);
av_opt_set_sample_fmt(swr, "in_sample_fmt", codec->sample_fmt, 0);
/* Convert it into 16khz Mono */
av_opt_set_chlayout(swr, "out_chlayout", &out_ch_layout, 0);
av_opt_set_int(swr, "out_sample_rate", WAVE_SAMPLE_RATE, 0);
av_opt_set_sample_fmt(swr, "out_sample_fmt", AV_SAMPLE_FMT_S16, 0);
#else
av_opt_set_int(swr, "in_channel_count", codec->channels, 0); av_opt_set_int(swr, "in_channel_count", codec->channels, 0);
av_opt_set_int(swr, "out_channel_count", 1, 0); av_opt_set_int(swr, "out_channel_count", 1, 0);
av_opt_set_int(swr, "in_channel_layout", codec->channel_layout, 0); av_opt_set_int(swr, "in_channel_layout", codec->channel_layout, 0);
@ -257,6 +271,7 @@ static int decode_audio(struct audio_buffer *audio_buf, s16 **data, int *size)
av_opt_set_int(swr, "out_sample_rate", WAVE_SAMPLE_RATE, 0); av_opt_set_int(swr, "out_sample_rate", WAVE_SAMPLE_RATE, 0);
av_opt_set_sample_fmt(swr, "in_sample_fmt", codec->sample_fmt, 0); av_opt_set_sample_fmt(swr, "in_sample_fmt", codec->sample_fmt, 0);
av_opt_set_sample_fmt(swr, "out_sample_fmt", AV_SAMPLE_FMT_S16, 0); av_opt_set_sample_fmt(swr, "out_sample_fmt", AV_SAMPLE_FMT_S16, 0);
#endif
swr_init(swr); swr_init(swr);
if (!swr_is_initialized(swr)) { if (!swr_is_initialized(swr)) {