ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-06-19 07:18:07 +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

10
bindings/java/src/main/java/io/github/ggerganov/whispercpp/WhisperCppJnaLibrary.java
View File

@ -20,7 +20,7 @@ public interface WhisperCppJnaLibrary extends Library {
* @return Whisper context on success, null on failure
*/
Pointer whisper_init_from_file(String path_model);
/**
* Provides default params which can be used with `whisper_init_from_file_with_params()` etc.
* Because this function allocates memory for the params, the caller must call either:
@ -304,14 +304,6 @@ public interface WhisperCppJnaLibrary extends Library {
/** Language id associated with the provided state */
int whisper_full_lang_id_from_state(Pointer state);
/**
* 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.
* @return 0 on success
*/
int whisper_pcm_to_mel_phase_vocoder(Pointer ctx, final float[] samples, int n_samples, int n_threads);
int whisper_pcm_to_mel_phase_vocoder_with_state(Pointer ctx, Pointer state, final float[] samples, int n_samples, int n_threads);
/** Get the start time of the specified segment. */
long whisper_full_get_segment_t0(Pointer ctx, int i_segment);

10
bindings/java/src/main/java/io/github/ggerganov/whispercpp/params/WhisperFullParams.java
View File

@ -129,14 +129,6 @@ public class WhisperFullParams extends Structure {
/** Maximum tokens per segment (0, default = no limit) */
public int max_tokens;
/** Flag to speed up the audio by 2x using Phase Vocoder. (default = false) */
public CBool speed_up;
/** Flag to speed up the audio by 2x using Phase Vocoder. (default = false) */
public void speedUp(boolean enable) {
speed_up = enable ? CBool.TRUE : CBool.FALSE;
}
/** Overwrite the audio context size (0 = use default). */
public int audio_ctx;
@ -321,7 +313,7 @@ public class WhisperFullParams extends Structure {
return Arrays.asList("strategy", "n_threads", "n_max_text_ctx", "offset_ms", "duration_ms", "translate",
"no_context", "single_segment", "no_timestamps",
"print_special", "print_progress", "print_realtime", "print_timestamps", "token_timestamps",
"thold_pt", "thold_ptsum", "max_len", "split_on_word", "max_tokens", "speed_up", "audio_ctx",
"thold_pt", "thold_ptsum", "max_len", "split_on_word", "max_tokens", "audio_ctx",
"tdrz_enable", "suppress_regex", "initial_prompt", "prompt_tokens", "prompt_n_tokens", "language", "detect_language",
"suppress_blank", "suppress_non_speech_tokens", "temperature", "max_initial_ts", "length_penalty",
"temperature_inc", "entropy_thold", "logprob_thold", "no_speech_thold", "greedy", "beam_search",