Spaces:
Sleeping
Sleeping
File size: 1,816 Bytes
a28f35e |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 |
#include "whisper.h"
#include "common-whisper.h"
#include <cstdio>
#include <cfloat>
#include <string>
#include <cstring>
#ifdef NDEBUG
#undef NDEBUG
#endif
#include <cassert>
int main() {
std::string whisper_model_path = "../../models/ggml-base.en.bin";
std::string vad_model_path = "../../models/for-tests-silero-v5.1.2-ggml.bin";
std::string sample_path = "../../samples/jfk.wav";
// Load the sample audio file
std::vector<float> pcmf32;
std::vector<std::vector<float>> pcmf32s;
assert(read_audio_data(sample_path.c_str(), pcmf32, pcmf32s, false));
struct whisper_context_params cparams = whisper_context_default_params();
struct whisper_context * wctx = whisper_init_from_file_with_params(
whisper_model_path.c_str(),
cparams);
struct whisper_full_params wparams = whisper_full_default_params(WHISPER_SAMPLING_BEAM_SEARCH);
wparams.vad = true;
wparams.vad_model_path = vad_model_path.c_str();
wparams.vad_params.threshold = 0.5f;
wparams.vad_params.min_speech_duration_ms = 250;
wparams.vad_params.min_silence_duration_ms = 100;
wparams.vad_params.max_speech_duration_s = FLT_MAX;
wparams.vad_params.speech_pad_ms = 30;
assert(whisper_full_parallel(wctx, wparams, pcmf32.data(), pcmf32.size(), 1) == 0);
const int n_segments = whisper_full_n_segments(wctx);
assert(n_segments == 1);
assert(strcmp(" And so my fellow Americans, ask not what your country can do for you,"
" ask what you can do for your country.",
whisper_full_get_segment_text(wctx, 0)) == 0);
assert(whisper_full_get_segment_t0(wctx, 0) == 29);
assert(whisper_full_get_segment_t1(wctx, 0) == 1049);
whisper_free(wctx);
return 0;
}
|