ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-06-13 04:28:07 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

whisper : support speaker segmentation (local diarization) of mono audio via tinydiarize (#1058)

Browse Source 
* add HuggingFace mirror to download  ggml model

* support tdrz via simple hack overriding solm tokens

* fix incorrect translate/transcribe token_ids that are not static const

* add apollo 13 sample for tdrz demo

* render [SPEAKER TURN] consistently in all terminal output using vocab.id_to_token

* extend whisper_segment with speaker_turn_next field and save in json output

* fix failing go build

* slipped in some python syntax whoops

* whisper : finalize tinydiarize support (add flag + fixes)

* whisper : tdrz support for word-level timestamps (respect max_len)

* java : try to fix tests after adding tdrz_enable flag

* main : remove TODO leftover

* java : fix params order list after adding "tdrz_enable"

* whisper : fix solm and add nosp token

* main : print tinydiarize help

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit is contained in:
Akash Mahajan
2023-07-03 23:45:00 -07:00
committed by GitHub
parent fdf58a6668
commit c8d0f5fe98
8 changed files with 215 additions and 130 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
bindings/java/src/main/java/io/github/ggerganov/whispercpp/WhisperCppJnaLibrary.java
View File

@ -224,8 +224,8 @@ public interface WhisperCppJnaLibrary extends Library {
int whisper_token_lang(Pointer ctx, int lang_id);
// Task tokens
int whisper_token_translate();
int whisper_token_transcribe();
int whisper_token_translate (Pointer ctx);
int whisper_token_transcribe(Pointer ctx);
// Performance information from the default state.
void whisper_print_timings(Pointer ctx);

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

@ -137,6 +137,14 @@ public class WhisperFullParams extends Structure {
/** Overwrite the audio context size (0 = use default). */
public int audio_ctx;
/** Enable tinydiarize (default = false) */
public CBool tdrz_enable;
/** Enable tinydiarize (default = false) */
public void tdrzEnable(boolean enable) {
tdrz_enable = enable ? CBool.TRUE : CBool.FALSE;
}
/** Tokens to provide to the whisper decoder as an initial prompt.
* These are prepended to any existing text context from a previous call. */
public String initial_prompt;
@ -302,7 +310,7 @@ public class WhisperFullParams extends Structure {
"no_context", "single_segment",
"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",
"initial_prompt", "prompt_tokens", "prompt_n_tokens", "language", "detect_language",
"tdrz_enable", "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",
"new_segment_callback", "new_segment_callback_user_data",