forked from extern/whisper.cpp
whisper : add whisper_state + default state on the whisper_context (#523)
* Added whisper state + default state on the whisper_context * Fixed some examples and bindings * Fixed whisper_n_len (which was used in some binding) and added whisper_n_len_from_state * Fixed comments * whisper : reuse kv_cache_free() and fix compiler warnings * whisper : clean-up the API comments --------- Co-authored-by: Sandro Hanea <sandrohanea@microsoft.com> Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit is contained in:
sandrohanea
GitHub
parent
478289a4b3
commit
59fdcd19c8
@ -20,7 +20,7 @@ extern bool callEncoderBegin(void* user_data);
|
||||
// Text segment callback
|
||||
// Called on every newly generated text segment
|
||||
// Use the whisper_full_...() functions to obtain the text segments
|
||||
static void whisper_new_segment_cb(struct whisper_context* ctx, int n_new, void* user_data) {
|
||||
static void whisper_new_segment_cb(struct whisper_context* ctx, struct whisper_state* state, int n_new, void* user_data) {
|
||||
if(user_data != NULL && ctx != NULL) {
|
||||
callNewSegment(user_data, n_new);
|
||||
}
|
||||
@ -29,7 +29,7 @@ static void whisper_new_segment_cb(struct whisper_context* ctx, int n_new, void*
|
||||
// Encoder begin callback
|
||||
// If not NULL, called before the encoder starts
|
||||
// If it returns false, the computation is aborted
|
||||
static bool whisper_encoder_begin_cb(struct whisper_context* ctx, void* user_data) {
|
||||
static bool whisper_encoder_begin_cb(struct whisper_context* ctx, struct whisper_state* state, void* user_data) {
|
||||
if(user_data != NULL && ctx != NULL) {
|
||||
return callEncoderBegin(user_data);
|
||||
}
|
||||
|
||||
@ -199,7 +199,7 @@ static VALUE ruby_whisper_transcribe(int argc, VALUE *argv, VALUE self) {
|
||||
{
|
||||
static bool is_aborted = false; // NOTE: this should be atomic to avoid data race
|
||||
|
||||
rwp->params.encoder_begin_callback = [](struct whisper_context * /*ctx*/, void * user_data) {
|
||||
rwp->params.encoder_begin_callback = [](struct whisper_context * /*ctx*/, struct whisper_state * /*state*/, void * user_data) {
|
||||
bool is_aborted = *(bool*)user_data;
|
||||
return !is_aborted;
|
||||
};
|
||||
|
||||
@ -72,7 +72,7 @@ int timestamp_to_sample(int64_t t, int n_samples) {
|
||||
return std::max(0, std::min((int) n_samples - 1, (int) ((t*WHISPER_SAMPLE_RATE)/100)));
|
||||
}
|
||||
|
||||
void whisper_print_segment_callback(struct whisper_context * ctx, int n_new, void * user_data) {
|
||||
void whisper_print_segment_callback(struct whisper_context * ctx, struct whisper_state * state, int n_new, void * user_data) {
|
||||
const auto & params = *((whisper_print_user_data *) user_data)->params;
|
||||
const auto & pcmf32s = *((whisper_print_user_data *) user_data)->pcmf32s;
|
||||
|
||||
@ -260,7 +260,7 @@ int run(whisper_params ¶ms, std::vector<std::vector<std::string>> &result) {
|
||||
{
|
||||
static bool is_aborted = false; // NOTE: this should be atomic to avoid data race
|
||||
|
||||
wparams.encoder_begin_callback = [](struct whisper_context * /*ctx*/, void * user_data) {
|
||||
wparams.encoder_begin_callback = [](struct whisper_context * /*ctx*/, struct whisper_state * /*state*/, void * user_data) {
|
||||
bool is_aborted = *(bool*)user_data;
|
||||
return !is_aborted;
|
||||
};
|
||||
|
||||
@ -193,7 +193,7 @@ struct whisper_print_user_data {
|
||||
const std::vector<std::vector<float>> * pcmf32s;
|
||||
};
|
||||
|
||||
void whisper_print_segment_callback(struct whisper_context * ctx, int n_new, void * user_data) {
|
||||
void whisper_print_segment_callback(struct whisper_context * ctx, struct whisper_state * /*state*/, int n_new, void * user_data) {
|
||||
const auto & params = *((whisper_print_user_data *) user_data)->params;
|
||||
const auto & pcmf32s = *((whisper_print_user_data *) user_data)->pcmf32s;
|
||||
|
||||
@ -608,7 +608,7 @@ int main(int argc, char ** argv) {
|
||||
{
|
||||
static bool is_aborted = false; // NOTE: this should be atomic to avoid data race
|
||||
|
||||
wparams.encoder_begin_callback = [](struct whisper_context * /*ctx*/, void * user_data) {
|
||||
wparams.encoder_begin_callback = [](struct whisper_context * /*ctx*/, struct whisper_state * /*state*/, void * user_data) {
|
||||
bool is_aborted = *(bool*)user_data;
|
||||
return !is_aborted;
|
||||
};
|
||||
|
||||
994
whisper.cpp
994
whisper.cpp
File diff suppressed because it is too large
Load Diff
154
whisper.h
154
whisper.h
@ -66,6 +66,7 @@ extern "C" {
|
||||
//
|
||||
|
||||
struct whisper_context;
|
||||
struct whisper_state;
|
||||
|
||||
typedef int whisper_token;
|
||||
|
||||
@ -101,11 +102,20 @@ extern "C" {
|
||||
WHISPER_API struct whisper_context * whisper_init_from_buffer(void * buffer, size_t buffer_size);
|
||||
WHISPER_API struct whisper_context * whisper_init(struct whisper_model_loader * loader);
|
||||
|
||||
// Frees all memory allocated by the model.
|
||||
WHISPER_API void whisper_free(struct whisper_context * ctx);
|
||||
// These are the same as the above, but the internal state of the context is not allocated automatically
|
||||
// It is the responsibility of the caller to allocate the state using whisper_init_state() (#523)
|
||||
WHISPER_API struct whisper_context * whisper_init_from_file_no_state(const char * path_model);
|
||||
WHISPER_API struct whisper_context * whisper_init_from_buffer_no_state(void * buffer, size_t buffer_size);
|
||||
WHISPER_API struct whisper_context * whisper_init_no_state(struct whisper_model_loader * loader);
|
||||
|
||||
WHISPER_API struct whisper_state * whisper_init_state(struct whisper_context * ctx);
|
||||
|
||||
// Frees all allocated memory
|
||||
WHISPER_API void whisper_free (struct whisper_context * ctx);
|
||||
WHISPER_API void whisper_free_state(struct whisper_state * state);
|
||||
|
||||
// Convert RAW PCM audio to log mel spectrogram.
|
||||
// The resulting spectrogram is stored inside the provided whisper context.
|
||||
// The resulting spectrogram is stored inside the default state of the provided whisper context.
|
||||
// Returns 0 on success
|
||||
WHISPER_API int whisper_pcm_to_mel(
|
||||
struct whisper_context * ctx,
|
||||
@ -113,17 +123,30 @@ extern "C" {
|
||||
int n_samples,
|
||||
int n_threads);
|
||||
|
||||
// Convert RAW PCM audio to log mel spectrogram but applies a Phase Vocoder to speed up the audio x2.
|
||||
// The resulting spectrogram is stored inside the provided whisper context.
|
||||
WHISPER_API int whisper_pcm_to_mel_with_state(
|
||||
struct whisper_context * ctx,
|
||||
struct whisper_state * state,
|
||||
const float * samples,
|
||||
int n_samples,
|
||||
int n_threads);
|
||||
|
||||
// Convert RAW PCM audio to log mel spectrogram but applies a Phase Vocoder to speed up the audio x2.
|
||||
// The resulting spectrogram is stored inside the default state of the provided whisper context.
|
||||
// Returns 0 on success
|
||||
WHISPER_API int whisper_pcm_to_mel_phase_vocoder(
|
||||
struct whisper_context* ctx,
|
||||
const float* samples,
|
||||
int n_samples,
|
||||
int n_threads);
|
||||
struct whisper_context * ctx,
|
||||
const float * samples,
|
||||
int n_samples,
|
||||
int n_threads);
|
||||
|
||||
WHISPER_API int whisper_pcm_to_mel_phase_vocoder_with_state(
|
||||
struct whisper_context * ctx,
|
||||
struct whisper_state * state,
|
||||
const float * samples,
|
||||
int n_samples,
|
||||
int n_threads);
|
||||
|
||||
// This can be used to set a custom log mel spectrogram inside the provided whisper context.
|
||||
// This can be used to set a custom log mel spectrogram inside the default state of the provided whisper context.
|
||||
// Use this instead of whisper_pcm_to_mel() if you want to provide your own log mel spectrogram.
|
||||
// n_mel must be 80
|
||||
// Returns 0 on success
|
||||
@ -133,7 +156,14 @@ extern "C" {
|
||||
int n_len,
|
||||
int n_mel);
|
||||
|
||||
// Run the Whisper encoder on the log mel spectrogram stored inside the provided whisper context.
|
||||
WHISPER_API int whisper_set_mel_with_state(
|
||||
struct whisper_context * ctx,
|
||||
struct whisper_state * state,
|
||||
const float * data,
|
||||
int n_len,
|
||||
int n_mel);
|
||||
|
||||
// Run the Whisper encoder on the log mel spectrogram stored inside the default state in the provided whisper context.
|
||||
// Make sure to call whisper_pcm_to_mel() or whisper_set_mel() first.
|
||||
// offset can be used to specify the offset of the first frame in the spectrogram.
|
||||
// Returns 0 on success
|
||||
@ -142,6 +172,12 @@ extern "C" {
|
||||
int offset,
|
||||
int n_threads);
|
||||
|
||||
WHISPER_API int whisper_encode_with_state(
|
||||
struct whisper_context * ctx,
|
||||
struct whisper_state * state,
|
||||
int offset,
|
||||
int n_threads);
|
||||
|
||||
// Run the Whisper decoder to obtain the logits and probabilities for the next token.
|
||||
// Make sure to call whisper_encode() first.
|
||||
// tokens + n_tokens is the provided context for the decoder.
|
||||
@ -155,6 +191,14 @@ extern "C" {
|
||||
int n_past,
|
||||
int n_threads);
|
||||
|
||||
WHISPER_API int whisper_decode_with_state(
|
||||
struct whisper_context * ctx,
|
||||
struct whisper_state * state,
|
||||
const whisper_token * tokens,
|
||||
int n_tokens,
|
||||
int n_past,
|
||||
int n_threads);
|
||||
|
||||
// Convert the provided text into tokens.
|
||||
// The tokens pointer must be large enough to hold the resulting tokens.
|
||||
// Returns the number of tokens on success, no more than n_max_tokens
|
||||
@ -190,17 +234,26 @@ extern "C" {
|
||||
int n_threads,
|
||||
float * lang_probs);
|
||||
|
||||
WHISPER_API int whisper_n_len (struct whisper_context * ctx); // mel length
|
||||
WHISPER_API int whisper_n_vocab (struct whisper_context * ctx);
|
||||
WHISPER_API int whisper_n_text_ctx (struct whisper_context * ctx);
|
||||
WHISPER_API int whisper_n_audio_ctx (struct whisper_context * ctx);
|
||||
WHISPER_API int whisper_is_multilingual(struct whisper_context * ctx);
|
||||
WHISPER_API int whisper_lang_auto_detect_with_state(
|
||||
struct whisper_context * ctx,
|
||||
struct whisper_state * state,
|
||||
int offset_ms,
|
||||
int n_threads,
|
||||
float * lang_probs);
|
||||
|
||||
WHISPER_API int whisper_n_len (struct whisper_context * ctx); // mel length
|
||||
WHISPER_API int whisper_n_len_from_state(struct whisper_state * state); // mel length
|
||||
WHISPER_API int whisper_n_vocab (struct whisper_context * ctx);
|
||||
WHISPER_API int whisper_n_text_ctx (struct whisper_context * ctx);
|
||||
WHISPER_API int whisper_n_audio_ctx (struct whisper_context * ctx);
|
||||
WHISPER_API int whisper_is_multilingual (struct whisper_context * ctx);
|
||||
|
||||
// Token logits obtained from the last call to whisper_decode()
|
||||
// The logits for the last token are stored in the last row
|
||||
// Rows: n_tokens
|
||||
// Cols: n_vocab
|
||||
WHISPER_API float * whisper_get_logits(struct whisper_context * ctx);
|
||||
WHISPER_API float * whisper_get_logits (struct whisper_context * ctx);
|
||||
WHISPER_API float * whisper_get_logits_from_state(struct whisper_state * state);
|
||||
|
||||
// Token Id -> String. Uses the vocabulary in the provided context
|
||||
WHISPER_API const char * whisper_token_to_str(struct whisper_context * ctx, whisper_token token);
|
||||
@ -218,7 +271,7 @@ extern "C" {
|
||||
WHISPER_API whisper_token whisper_token_translate (void);
|
||||
WHISPER_API whisper_token whisper_token_transcribe(void);
|
||||
|
||||
// Performance information
|
||||
// Performance information from the default state.
|
||||
WHISPER_API void whisper_print_timings(struct whisper_context * ctx);
|
||||
WHISPER_API void whisper_reset_timings(struct whisper_context * ctx);
|
||||
|
||||
@ -236,18 +289,19 @@ extern "C" {
|
||||
// Text segment callback
|
||||
// Called on every newly generated text segment
|
||||
// Use the whisper_full_...() functions to obtain the text segments
|
||||
typedef void (*whisper_new_segment_callback)(struct whisper_context * ctx, int n_new, void * user_data);
|
||||
typedef void (*whisper_new_segment_callback)(struct whisper_context * ctx, struct whisper_state * state, int n_new, void * user_data);
|
||||
|
||||
// Encoder begin callback
|
||||
// If not NULL, called before the encoder starts
|
||||
// If it returns false, the computation is aborted
|
||||
typedef bool (*whisper_encoder_begin_callback)(struct whisper_context * ctx, void * user_data);
|
||||
typedef bool (*whisper_encoder_begin_callback)(struct whisper_context * ctx, struct whisper_state * state, void * user_data);
|
||||
|
||||
// Logits filter callback
|
||||
// Can be used to modify the logits before sampling
|
||||
// If not NULL, called after applying temperature to logits
|
||||
typedef void (*whisper_logits_filter_callback)(
|
||||
struct whisper_context * ctx,
|
||||
struct whisper_state * state,
|
||||
const whisper_token_data * tokens,
|
||||
int n_tokens,
|
||||
float * logits,
|
||||
@ -334,6 +388,7 @@ extern "C" {
|
||||
WHISPER_API struct whisper_full_params whisper_full_default_params(enum whisper_sampling_strategy strategy);
|
||||
|
||||
// Run the entire model: PCM -> log mel spectrogram -> encoder -> decoder -> text
|
||||
// Not thread safe for same context
|
||||
// Uses the specified decoding strategy to obtain the text.
|
||||
WHISPER_API int whisper_full(
|
||||
struct whisper_context * ctx,
|
||||
@ -341,7 +396,16 @@ extern "C" {
|
||||
const float * samples,
|
||||
int n_samples);
|
||||
|
||||
// Split the input audio in chunks and process each chunk separately using whisper_full()
|
||||
WHISPER_API int whisper_full_with_state(
|
||||
struct whisper_context * ctx,
|
||||
struct whisper_state * state,
|
||||
struct whisper_full_params params,
|
||||
const float * samples,
|
||||
int n_samples);
|
||||
|
||||
// Split the input audio in chunks and process each chunk separately using whisper_full_with_state()
|
||||
// Result is stored in the default state of the context
|
||||
// Not thread safe if executed in parallel on the same context.
|
||||
// It seems this approach can offer some speedup in some cases.
|
||||
// However, the transcription accuracy can be worse at the beginning and end of each chunk.
|
||||
WHISPER_API int whisper_full_parallel(
|
||||
@ -351,33 +415,47 @@ extern "C" {
|
||||
int n_samples,
|
||||
int n_processors);
|
||||
|
||||
// Number of generated text segments.
|
||||
// Number of generated text segments
|
||||
// A segment can be a few words, a sentence, or even a paragraph.
|
||||
WHISPER_API int whisper_full_n_segments(struct whisper_context * ctx);
|
||||
WHISPER_API int whisper_full_n_segments (struct whisper_context * ctx);
|
||||
WHISPER_API int whisper_full_n_segments_from_state(struct whisper_state * state);
|
||||
|
||||
// Language id associated with the current context
|
||||
// Language id associated with the context's default state
|
||||
WHISPER_API int whisper_full_lang_id(struct whisper_context * ctx);
|
||||
|
||||
// Get the start and end time of the specified segment.
|
||||
WHISPER_API int64_t whisper_full_get_segment_t0(struct whisper_context * ctx, int i_segment);
|
||||
WHISPER_API int64_t whisper_full_get_segment_t1(struct whisper_context * ctx, int i_segment);
|
||||
// Language id associated with the provided state
|
||||
WHISPER_API int whisper_full_lang_id_from_state(struct whisper_state * state);
|
||||
|
||||
// Get the text of the specified segment.
|
||||
WHISPER_API const char * whisper_full_get_segment_text(struct whisper_context * ctx, int i_segment);
|
||||
// Get the start and end time of the specified segment
|
||||
WHISPER_API int64_t whisper_full_get_segment_t0 (struct whisper_context * ctx, int i_segment);
|
||||
WHISPER_API int64_t whisper_full_get_segment_t0_from_state(struct whisper_state * state, int i_segment);
|
||||
|
||||
// Get number of tokens in the specified segment.
|
||||
WHISPER_API int whisper_full_n_tokens(struct whisper_context * ctx, int i_segment);
|
||||
WHISPER_API int64_t whisper_full_get_segment_t1 (struct whisper_context * ctx, int i_segment);
|
||||
WHISPER_API int64_t whisper_full_get_segment_t1_from_state(struct whisper_state * state, int i_segment);
|
||||
|
||||
// Get the token text of the specified token in the specified segment.
|
||||
WHISPER_API const char * whisper_full_get_token_text(struct whisper_context * ctx, int i_segment, int i_token);
|
||||
WHISPER_API whisper_token whisper_full_get_token_id (struct whisper_context * ctx, int i_segment, int i_token);
|
||||
// Get the text of the specified segment
|
||||
WHISPER_API const char * whisper_full_get_segment_text (struct whisper_context * ctx, int i_segment);
|
||||
WHISPER_API const char * whisper_full_get_segment_text_from_state(struct whisper_state * state, int i_segment);
|
||||
|
||||
// Get token data for the specified token in the specified segment.
|
||||
// Get number of tokens in the specified segment
|
||||
WHISPER_API int whisper_full_n_tokens (struct whisper_context * ctx, int i_segment);
|
||||
WHISPER_API int whisper_full_n_tokens_from_state(struct whisper_state * state, int i_segment);
|
||||
|
||||
// Get the token text of the specified token in the specified segment
|
||||
WHISPER_API const char * whisper_full_get_token_text (struct whisper_context * ctx, int i_segment, int i_token);
|
||||
WHISPER_API const char * whisper_full_get_token_text_from_state(struct whisper_context * ctx, struct whisper_state * state, int i_segment, int i_token);
|
||||
|
||||
WHISPER_API whisper_token whisper_full_get_token_id (struct whisper_context * ctx, int i_segment, int i_token);
|
||||
WHISPER_API whisper_token whisper_full_get_token_id_from_state(struct whisper_state * state, int i_segment, int i_token);
|
||||
|
||||
// Get token data for the specified token in the specified segment
|
||||
// This contains probabilities, timestamps, etc.
|
||||
WHISPER_API whisper_token_data whisper_full_get_token_data(struct whisper_context * ctx, int i_segment, int i_token);
|
||||
WHISPER_API whisper_token_data whisper_full_get_token_data (struct whisper_context * ctx, int i_segment, int i_token);
|
||||
WHISPER_API whisper_token_data whisper_full_get_token_data_from_state(struct whisper_state * state, int i_segment, int i_token);
|
||||
|
||||
// Get the probability of the specified token in the specified segment.
|
||||
WHISPER_API float whisper_full_get_token_p(struct whisper_context * ctx, int i_segment, int i_token);
|
||||
// Get the probability of the specified token in the specified segment
|
||||
WHISPER_API float whisper_full_get_token_p (struct whisper_context * ctx, int i_segment, int i_token);
|
||||
WHISPER_API float whisper_full_get_token_p_from_state(struct whisper_state * state, int i_segment, int i_token);
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////
|
||||
|
||||
|
||||
Loading…
Reference in New Issue
Block a user