LLM-Inference/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/whisper.cpp.git synced 2026-08-12 22:31:22 +04:00
Code Issues Packages Projects Releases Wiki Activity

examples : fix argument flag for min speech duration in VAD (#3907)

Browse Source 
Fixed the -vspd flag for vad_min_speech_duration_ms, to prevent hiding vad_min_silence_duration_ms.
In usage () clarified the output timestamp units.
Fixed a few typos.
This commit is contained in:
QuantiusBenignus
2026-06-26 08:09:03 +02:00
committed by GitHub
parent 43d78af5be
commit e3ab311094
1 changed files with 3 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
examples/vad-speech-segments/speech.cpp
+3 -2
View File
@@ -26,6 +26,7 @@ static void vad_print_usage(int /*argc*/, char ** argv, const cli_params & param
fprintf(stderr, "\n");
fprintf(stderr, "usage: %s [options] file\n", argv[0]);
fprintf(stderr, "supported audio formats: flac, mp3, ogg, wav\n");
fprintf(stderr, "Note: Output timestamps are in centiseconds (1/100th of a second).\n");
fprintf(stderr, "\n");
fprintf(stderr, "options:\n");
fprintf(stderr, " -h, --help [default] show this help message and exit\n");
@@ -64,7 +65,7 @@ static bool vad_params_parse(int argc, char ** argv, cli_params & params) {
else if (arg == "-ug" || arg == "--use-gpu") { params.use_gpu = true; }
else if (arg == "-vm" || arg == "--vad-model") { params.vad_model = ARGV_NEXT; }
else if (arg == "-vt" || arg == "--vad-threshold") { params.vad_threshold = std::stof(ARGV_NEXT); }
else if (arg == "-vsd" || arg == "--vad-min-speech-duration-ms") { params.vad_min_speech_duration_ms = std::stoi(ARGV_NEXT); }
else if (arg == "-vspd" || arg == "--vad-min-speech-duration-ms") { params.vad_min_speech_duration_ms = std::stoi(ARGV_NEXT); }
else if (arg == "-vsd" || arg == "--vad-min-silence-duration-ms") { params.vad_min_speech_duration_ms = std::stoi(ARGV_NEXT); }
else if (arg == "-vmsd" || arg == "--vad-max-speech-duration-s") { params.vad_max_speech_duration_s = std::stof(ARGV_NEXT); }
else if (arg == "-vp" || arg == "--vad-speech-pad-ms") { params.vad_speech_pad_ms = std::stoi(ARGV_NEXT); }
@@ -122,7 +123,7 @@ int main(int argc, char ** argv) {
return 3;
}
// Get the the vad segements using the probabilities that have been computed
// Get the the vad segments using the probabilities that have been computed
// previously and stored in the whisper_vad_context.
struct whisper_vad_params params = whisper_vad_default_params();
params.threshold = cli_params.vad_threshold;