ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-01-01 18:26:40 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

whisper : add support for new distilled Whisper models

Browse Source
This commit is contained in:
Georgi Gerganov 2023-11-03 13:24:05 +02:00
parent 8a2bee6717
commit b8c93c5f3b
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
1 changed files with 12 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
whisper.cpp
View File

@ -3942,6 +3942,7 @@ static void whisper_process_logits(
// suppress task tokens // suppress task tokens
logits[vocab.token_translate] = -INFINITY; logits[vocab.token_translate] = -INFINITY;
logits[vocab.token_transcribe] = -INFINITY; logits[vocab.token_transcribe] = -INFINITY;
logits[vocab.token_prev] = -INFINITY;
if (params.logits_filter_callback) { if (params.logits_filter_callback) {
params.logits_filter_callback(&ctx, &state, tokens_cur.data(), tokens_cur.size(), logits.data(), params.logits_filter_callback_user_data); params.logits_filter_callback(&ctx, &state, tokens_cur.data(), tokens_cur.size(), logits.data(), params.logits_filter_callback_user_data);
@ -4560,6 +4561,7 @@ int whisper_full_with_state(
// these tokens determine the task that will be performed // these tokens determine the task that will be performed
std::vector<whisper_token> prompt_init = { whisper_token_sot(ctx) }; std::vector<whisper_token> prompt_init = { whisper_token_sot(ctx) };
if (whisper_is_multilingual(ctx)) { if (whisper_is_multilingual(ctx)) {
const int lang_id = whisper_lang_id(params.language); const int lang_id = whisper_lang_id(params.language);
state->lang_id = lang_id; state->lang_id = lang_id;
@ -4571,6 +4573,16 @@ int whisper_full_with_state(
} }
} }
{
const bool is_distil = ctx->model.hparams.n_text_layer == 2;
// distilled models require the "no_timestamps" token
// TODO: add input parameter (#1229)
if (is_distil) {
prompt_init.push_back(whisper_token_not(ctx));
}
}
int seek = seek_start; int seek = seek_start;
std::vector<whisper_token> prompt; std::vector<whisper_token> prompt;