extern/whisper.cpp
1
0
Fork 1
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-01-01 03:28:57 +01:00
Code Issues Packages Projects Releases Wiki Activity

main : make whisper_print_segment_callback() more readable (close #371)

Browse Source
This commit is contained in:
Georgi Gerganov 2023-01-05 21:45:05 +02:00
parent 4a214d2f07
commit a0d4f8e65c
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
1 changed files with 68 additions and 76 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

46
examples/main/main.cpp
View File

@ -176,42 +176,30 @@ void whisper_print_segment_callback(struct whisper_context * ctx, int n_new, voi
const int n_segments = whisper_full_n_segments(ctx);
std::string speaker = "";
int64_t t0;
int64_t t1;
// print the last n_new segments
const int s0 = n_segments - n_new;
if (s0 == 0) {
printf("\n");
}
for (int i = s0; i < n_segments; i++) {
if (params.no_timestamps) {
if (params.print_colors) {
for (int j = 0; j < whisper_full_n_tokens(ctx, i); ++j) {
if (params.print_special == false) {
const whisper_token id = whisper_full_get_token_id(ctx, i, j);
if (id >= whisper_token_eot(ctx)) {
continue;
}
if (!params.no_timestamps || params.diarize) {
t0 = whisper_full_get_segment_t0(ctx, i);
t1 = whisper_full_get_segment_t1(ctx, i);
}
const char * text = whisper_full_get_token_text(ctx, i, j);
const float p = whisper_full_get_token_p (ctx, i, j);
const int col = std::max(0, std::min((int) k_colors.size(), (int) (std::pow(p, 3)*float(k_colors.size()))));
printf("%s%s%s", k_colors[col].c_str(), text, "\033[0m");
if (!params.no_timestamps) {
printf("[%s --> %s] ", to_timestamp(t0).c_str(), to_timestamp(t1).c_str());
}
} else {
const char * text = whisper_full_get_segment_text(ctx, i);
printf("%s", text);
}
fflush(stdout);
} else {
const int64_t t0 = whisper_full_get_segment_t0(ctx, i);
const int64_t t1 = whisper_full_get_segment_t1(ctx, i);
std::string speaker;
if (params.diarize && pcmf32s.size() == 2) {
const int64_t n_samples = pcmf32s[0].size();
const int64_t is0 = timestamp_to_sample(t0, n_samples);
@ -237,7 +225,6 @@ void whisper_print_segment_callback(struct whisper_context * ctx, int n_new, voi
}
if (params.print_colors) {
printf("[%s --> %s] ", to_timestamp(t0).c_str(), to_timestamp(t1).c_str());
for (int j = 0; j < whisper_full_n_tokens(ctx, i); ++j) {
if (params.print_special == false) {
const whisper_token id = whisper_full_get_token_id(ctx, i, j);
@ -253,13 +240,18 @@ void whisper_print_segment_callback(struct whisper_context * ctx, int n_new, voi
printf("%s%s%s%s", speaker.c_str(), k_colors[col].c_str(), text, "\033[0m");
}
printf("\n");
} else {
const char * text = whisper_full_get_segment_text(ctx, i);
printf("[%s --> %s] %s%s\n", to_timestamp(t0).c_str(), to_timestamp(t1).c_str(), speaker.c_str(), text);
printf("%s%s", speaker.c_str(), text);
}
// with timestamps or speakers: each segment on new line
if (!params.no_timestamps || params.diarize) {
printf("\n");
}
fflush(stdout);
}
}