ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-06-14 12:58:10 +00:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity

whisper : remove speed_up and phase_vocoder* functions (#2198)

Browse Source 
* whisper : fix cast warning

* whisper : remove phase_vocoder functions, ref #2195

* whisper : remove speed_up from whisper_full_params, closes #2195
This commit is contained in:
Borislav Stanimirov
2024-05-31 11:37:29 +03:00
committed by GitHub
parent b87494bb8f
commit af5833e298
20 changed files with 14 additions and 161 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
examples/lsp/lsp.cpp
View File

@ -26,7 +26,6 @@ struct whisper_params {
float vad_thold = 0.6f;
float freq_thold = 100.0f;
bool speed_up = false;
bool translate = false;
bool print_special = false;
bool print_energy = false;
@ -70,7 +69,6 @@ bool whisper_params_parse(int argc, char ** argv, whisper_params & params) {
else if (arg == "-ac" || arg == "--audio-ctx") { params.audio_ctx = std::stoi(argv[++i]); }
else if (arg == "-vth" || arg == "--vad-thold") { params.vad_thold = std::stof(argv[++i]); }
else if (arg == "-fth" || arg == "--freq-thold") { params.freq_thold = std::stof(argv[++i]); }
else if (arg == "-su" || arg == "--speed-up") { params.speed_up = true; }
else if (arg == "-tr" || arg == "--translate") { params.translate = true; }
else if (arg == "-ps" || arg == "--print-special") { params.print_special = true; }
else if (arg == "-pe" || arg == "--print-energy") { params.print_energy = true; }
@ -102,7 +100,6 @@ void whisper_print_usage(int /*argc*/, char ** argv, const whisper_params & para
fprintf(stderr, " -ac N, --audio-ctx N [%-7d] audio context size (0 - all)\n", params.audio_ctx);
fprintf(stderr, " -vth N, --vad-thold N [%-7.2f] voice activity detection threshold\n", params.vad_thold);
fprintf(stderr, " -fth N, --freq-thold N [%-7.2f] high-pass frequency cutoff\n", params.freq_thold);
fprintf(stderr, " -su, --speed-up [%-7s] speed up audio by x2 (reduced accuracy)\n", params.speed_up ? "true" : "false");
fprintf(stderr, " -tr, --translate [%-7s] translate from source language to english\n", params.translate ? "true" : "false");
fprintf(stderr, " -ps, --print-special [%-7s] print special tokens\n", params.print_special ? "true" : "false");
fprintf(stderr, " -pe, --print-energy [%-7s] print sound energy (for debugging)\n", params.print_energy ? "true" : "false");
@ -184,7 +181,6 @@ json unguided_transcription(struct whisper_context * ctx, audio_async &audio, js
wparams.n_threads = params.n_threads;
wparams.audio_ctx = params.audio_ctx;
wparams.speed_up = params.speed_up;
wparams.suppress_non_speech_tokens = true;
// run the transformer and a single decoding pass
if (whisper_full(ctx, wparams, pcmf32.data(), pcmf32.size()) != 0) {
@ -223,7 +219,6 @@ json guided_transcription(struct whisper_context * ctx, audio_async &audio, cons
wparams.n_threads = params.n_threads;
wparams.audio_ctx = params.audio_ctx;
wparams.speed_up = params.speed_up;
// TODO: Do some time testing. Does an overly long prompt slow down processing?
// Set up command sets/precompute prompts