ailia_speech  1.6.0.0
ailia_speech.h
[詳解]
1 
9 #ifndef INCLUDED_AILIA_SPEECH
10 #define INCLUDED_AILIA_SPEECH
11 
12 /* エラーコードと構造体 */
13 
14 #include "ailia.h"
15 #include "ailia_tokenizer.h"
16 
17 /* 呼び出し規約 */
18 
19 #if defined(_WIN64) || defined(_M_X64) || defined(__amd64__) || defined(__x86_64__) || defined(__APPLE__) || \
20  defined(__ANDROID__) || defined(ANDROID) || defined(__linux__) || defined(NN_NINTENDO_SDK)
21 #define AILIA_API
22 #else
23 #define AILIA_API __stdcall
24 #endif
25 
26 #ifdef __cplusplus
27 extern "C" {
28 #endif
29 
30 /****************************************************************
31  * モデルタイプ定義
32  **/
33 
43 #define AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_TINY (0)
44 
54 #define AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_BASE (1)
55 
65 #define AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_SMALL (2)
66 
76 #define AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_MEDIUM (3)
77 
87 #define AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_LARGE (4)
88 
98 #define AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_LARGE_V3 (5)
99 
109 #define AILIA_SPEECH_MODEL_TYPE_SENSEVOICE_SMALL (10)
110 
111 /****************************************************************
112  * タスク定義
113  **/
114 
124 #define AILIA_SPEECH_TASK_TRANSCRIBE (0)
125 
135 #define AILIA_SPEECH_TASK_TRANSLATE (1)
136 
137 /****************************************************************
138  * 制約定義
139  **/
140 
150 #define AILIA_SPEECH_CONSTRAINT_CHARACTERS (0)
151 
161 #define AILIA_SPEECH_CONSTRAINT_WORDS (1)
162 
163 /****************************************************************
164  * フラグ定義
165  **/
166 
176 #define AILIA_SPEECH_FLAG_NONE (0)
177 
189 #define AILIA_SPEECH_FLAG_LIVE (1)
190 
191 /****************************************************************
192  * VAD定義
193  **/
194 
204 #define AILIA_SPEECH_VAD_TYPE_SILERO (0)
205 
206 /****************************************************************
207  * DIARIZATION定義
208  **/
209 
219 #define AILIA_SPEECH_DIARIZATION_TYPE_PYANNOTE_AUDIO (0)
220 
221 /****************************************************************
222  * 辞書定義
223  **/
224 
234 #define AILIA_SPEECH_DICTIONARY_TYPE_REPLACE (0)
235 
236 /****************************************************************
237  * 後処理定義
238  **/
239 
249 #define AILIA_SPEECH_POST_PROCESS_TYPE_T5 (0)
250 
264 #define AILIA_SPEECH_POST_PROCESS_TYPE_FUGUMT_EN_JA (1)
265 
279 #define AILIA_SPEECH_POST_PROCESS_TYPE_FUGUMT_JA_EN (2)
280 
281 /****************************************************************
282  * APIコールバック定義
283  **/
284 
285 // 接続が必要なAPI
286 
287 #if defined(_WIN32) && !defined(_WIN64)
288 #define AILIA_SPEECH_USER_API __stdcall
289 #else
290 #define AILIA_SPEECH_USER_API
291 #endif
292 
293 // ailia.audio API
295 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_AUDIO_GET_MEL_SPECTROGRAM)(void*, const void*, int, int, int, int, int, int, int, int, float, int, float, float, int, int, int);
296 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_AUDIO_RESAMPLE)(void*, const void*, int, int, int, int);
298 
299 // ailia.tokenizer API
300 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_TOKENIZER_CREATE)(struct AILIATokenizer**, int, int);
301 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_TOKENIZER_OPEN_MODEL_FILE_A)(struct AILIATokenizer*, const char*);
302 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_TOKENIZER_OPEN_MODEL_FILE_W)(struct AILIATokenizer*, const wchar_t*);
303 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_TOKENIZER_ENCODE)(struct AILIATokenizer*, const char*);
304 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_TOKENIZER_GET_TOKEN_COUNT)(struct AILIATokenizer*, unsigned int*);
305 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_TOKENIZER_GET_TOKENS)(struct AILIATokenizer*, int*, unsigned int);
306 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_TOKENIZER_DECODE)(struct AILIATokenizer*, const int*, unsigned int);
307 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_TOKENIZER_GET_TEXT_LENGTH)(struct AILIATokenizer*, unsigned int*);
308 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_TOKENIZER_GET_TEXT)(struct AILIATokenizer*, char*, unsigned int);
310 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_TOKENIZER_UTF8_TO_UTF32)(unsigned int*, unsigned int*, const char*, unsigned int);
311 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_TOKENIZER_UTF32_TO_UTF8)(char*, unsigned int*, unsigned int);
312 
313 // ailia API
314 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_CREATE)(struct AILIANetwork**, int, int);
315 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_OPEN_WEIGHT_FILE_A)(struct AILIANetwork*, const char*);
316 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_OPEN_WEIGHT_FILE_W)(struct AILIANetwork*, const wchar_t*);
317 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_OPEN_WEIGHT_MEM)(struct AILIANetwork*, const void*, unsigned int);
318 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_SET_MEMORY_MODE)(struct AILIANetwork*, unsigned int);
319 typedef void(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_DESTROY)(struct AILIANetwork*);
320 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_UPDATE)(struct AILIANetwork*);
321 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_GET_BLOB_INDEX_BY_INPUT_INDEX)(struct AILIANetwork*, unsigned int*, unsigned int);
322 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_GET_BLOB_INDEX_BY_OUTPUT_INDEX)(struct AILIANetwork*, unsigned int*, unsigned int);
323 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_GET_BLOB_DATA)(struct AILIANetwork*, void*, unsigned int, unsigned int);
324 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_SET_INPUT_BLOB_DATA)(struct AILIANetwork*, const void*, unsigned int, unsigned int);
325 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_SET_INPUT_BLOB_SHAPE)(struct AILIANetwork*, const AILIAShape*, unsigned int, unsigned int);
326 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_GET_BLOB_SHAPE)(struct AILIANetwork*, AILIAShape*, unsigned int, unsigned int);
327 typedef const char*(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_GET_ERROR_DETAIL)(struct AILIANetwork*);
328 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_COPY_BLOB_DATA)(struct AILIANetwork* dst_net, unsigned int dst_blob_idx, struct AILIANetwork* src_net, unsigned int src_blob_idx);
329 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_AILIA_GET_ENVIRONMENT)(AILIAEnvironment** env, unsigned int env_idx, unsigned int version);
330 
340 #define AILIA_SPEECH_API_CALLBACK_VERSION (6)
341 
342 /* APIコールバック関数構造体 */
343 typedef struct _AILIASpeechApiCallback {
377 
378 /****************************************************************
379  * 中間情報取得コールバック
380  **/
381 
397 typedef int(AILIA_SPEECH_USER_API* AILIA_SPEECH_USER_API_INTERMEDIATE_CALLBACK)(void* handle, const char* text);
398 
399 /****************************************************************
400  * ネットワークオブジェクトのインスタンス
401  **/
402 
403 struct AILIASpeech;
404 
414 #define AILIA_SPEECH_TEXT_VERSION (2)
415 
425 #define AILIA_SPEECH_SPEAKER_ID_UNKNOWN 0xFFFFFFFF
426 
427 typedef struct _AILIASpeechText {
428  const char* text;
431  unsigned int speaker_id;
432  const char* language;
433  float confidence;
435 
436 /****************************************************************
437  * Speech2Text API
438  **/
439 
471 int AILIA_API ailiaSpeechCreate(struct AILIASpeech** net, int env_id, int num_thread, int memory_mode, int task, int flags, AILIASpeechApiCallback callback, int version);
472 
496 int AILIA_API ailiaSpeechOpenModelFileA(struct AILIASpeech* net, const char* encoder_path, const char* decoder_path, int model_type);
497 
525 int AILIA_API ailiaSpeechOpenModelFileW(struct AILIASpeech* net, const wchar_t* encoder_path, const wchar_t* decoder_path, int model_type);
526 
548 int AILIA_API ailiaSpeechOpenVadFileA(struct AILIASpeech* net, const char* vad_path, int vad_type);
549 
571 int AILIA_API ailiaSpeechOpenVadFileW(struct AILIASpeech* net, const wchar_t* vad_path, int vad_type);
572 
590 int AILIA_API ailiaSpeechOpenDictionaryFileA(struct AILIASpeech* net, const char* dictionary_path, int dictionary_type);
591 
609 int AILIA_API ailiaSpeechOpenDictionaryFileW(struct AILIASpeech* net, const wchar_t* dictionary_path, int dictionary_type);
610 
636 int AILIA_API ailiaSpeechOpenPostProcessFileA(struct AILIASpeech* net, const char* encoder_path, const char* decoder_path, const char* source_path, const char* target_path, const char* prefix, int post_process_type);
637 
663 int AILIA_API ailiaSpeechOpenPostProcessFileW(struct AILIASpeech* net, const wchar_t* encoder_path, const wchar_t* decoder_path, const wchar_t* source_path, const wchar_t* target_path, const char* prefix, int post_process_type);
664 
684 int AILIA_API ailiaSpeechOpenDiarizationFileA(struct AILIASpeech* net, const char* segmentation_path, const char* embedding_path, int type);
685 
705 int AILIA_API ailiaSpeechOpenDiarizationFileW(struct AILIASpeech* net, const wchar_t* segmentation_path, const wchar_t* embedding_path, int type);
706 
728 int AILIA_API
729 ailiaSpeechPushInputData(struct AILIASpeech* net, const float* src, unsigned int channels, unsigned int samples, unsigned int sampling_rate);
730 
750 int AILIA_API
751 ailiaSpeechFinalizeInputData(struct AILIASpeech* net);
752 
768 int AILIA_API ailiaSpeechBuffered(struct AILIASpeech* net, unsigned int* buffered);
769 
785 int AILIA_API
786 ailiaSpeechComplete(struct AILIASpeech* net, unsigned int* complete);
787 
803 int AILIA_API
804 ailiaSpeechSetPrompt(struct AILIASpeech* net, const char* prompt);
805 
823 int AILIA_API
824 ailiaSpeechSetConstraint(struct AILIASpeech* net, const char* constraint, int type);
825 
845 int AILIA_API
846 ailiaSpeechSetLanguage(struct AILIASpeech* net, const char* language);
847 
872 int AILIA_API
873 ailiaSpeechSetSilentThreshold(struct AILIASpeech* net, float silent_threshold, float speech_sec, float no_speech_sec);
874 
898 int AILIA_API
900 
918 int AILIA_API
919 ailiaSpeechTranscribe(struct AILIASpeech* net);
920 
938 int AILIA_API
939 ailiaSpeechPostProcess(struct AILIASpeech* net);
940 
956 int AILIA_API
957 ailiaSpeechGetTextCount(struct AILIASpeech* net, unsigned int* count);
958 
984 int AILIA_API
985 ailiaSpeechGetText(struct AILIASpeech* net, AILIASpeechText* text, unsigned int version, unsigned int idx);
986 
1014 int AILIA_API
1015 ailiaSpeechSetText(struct AILIASpeech* net, const AILIASpeechText* text, unsigned int version, unsigned int idx);
1016 
1026 void AILIA_API ailiaSpeechDestroy(struct AILIASpeech* net);
1027 
1049 int AILIA_API ailiaSpeechResetTranscribeState(struct AILIASpeech* net);
1050 
1072 const char* AILIA_API ailiaSpeechGetErrorDetail(struct AILIASpeech* net);
1073 
1074 #ifdef UNICODE
1075 #define ailiaSpeechOpenModelFile ailiaSpeechOpenModelFileW
1076 #define ailiaSpeechOpenVadFile ailiaSpeechOpenVadFileW
1077 #define ailiaSpeechOpenDiarizationFile ailiaSpeechOpenDiarizationFileW
1078 #define ailiaSpeechOpenDictionaryFile ailiaSpeechOpenDictionaryFileW
1079 #define ailiaSpeechOpenPostProcessFile ailiaSpeechOpenPostProcessFileW
1080 #else
1081 #define ailiaSpeechOpenModelFile ailiaSpeechOpenModelFileA
1082 #define ailiaSpeechOpenVadFile ailiaSpeechOpenVadFileA
1083 #define ailiaSpeechOpenDiarizationFile ailiaSpeechOpenDiarizationFileA
1084 #define ailiaSpeechOpenDictionaryFile ailiaSpeechOpenDictionaryFileA
1085 #define ailiaSpeechOpenPostProcessFile ailiaSpeechOpenPostProcessFileA
1086 #endif
1087 
1088 #ifdef __cplusplus
1089 }
1090 #endif
1091 #endif /* !defined(INCLUDED_AILIA_SPEECH) */
int AILIA_API ailiaSpeechOpenVadFileA(struct AILIASpeech *net, const char *vad_path, int vad_type)
無音検知に適用するVADモデルを指定します。(MBSC)
void AILIA_API ailiaSpeechDestroy(struct AILIASpeech *net)
ネットワークオブジェクトを破棄します。
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_SET_INPUT_BLOB_DATA)(struct AILIANetwork *, const void *, unsigned int, unsigned int)
Definition: ailia_speech.h:324
int AILIA_API ailiaSpeechBuffered(struct AILIASpeech *net, unsigned int *buffered)
音声認識を行うために十分なデータが存在するかどうかを判定します。
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_AUDIO_RESAMPLE)(void *, const void *, int, int, int, int)
Definition: ailia_speech.h:296
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_TOKENIZER_OPEN_MODEL_FILE_A)(struct AILIATokenizer *, const char *)
Definition: ailia_speech.h:301
int AILIA_API ailiaSpeechOpenVadFileW(struct AILIASpeech *net, const wchar_t *vad_path, int vad_type)
無音検知に適用するVADモデルを指定します。(UTF16)
#define AILIA_API
Definition: ailia_speech.h:23
int AILIA_API ailiaSpeechSetText(struct AILIASpeech *net, const AILIASpeechText *text, unsigned int version, unsigned int idx)
ポストプロセス対象のテキストを設定します。
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_TOKENIZER_GET_TOKEN_COUNT)(struct AILIATokenizer *, unsigned int *)
Definition: ailia_speech.h:304
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_GET_BLOB_SHAPE)(struct AILIANetwork *, AILIAShape *, unsigned int, unsigned int)
Definition: ailia_speech.h:326
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_TOKENIZER_ENCODE)(struct AILIATokenizer *, const char *)
Definition: ailia_speech.h:303
struct _AILIASpeechApiCallback AILIASpeechApiCallback
void(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_TOKENIZER_DESTROY)(struct AILIATokenizer *)
Definition: ailia_speech.h:309
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_CREATE)(struct AILIANetwork **, int, int)
Definition: ailia_speech.h:314
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_AUDIO_GET_RESAMPLE_LEN)(int *, int, int, int)
Definition: ailia_speech.h:297
int AILIA_API ailiaSpeechPushInputData(struct AILIASpeech *net, const float *src, unsigned int channels, unsigned int samples, unsigned int sampling_rate)
音声をキューに投入します。
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_AUDIO_GET_FRAME_LEN)(int *, int, int, int, int)
Definition: ailia_speech.h:294
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_INTERMEDIATE_CALLBACK)(void *handle, const char *text)
推論の途中の状況を通知します。
Definition: ailia_speech.h:397
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_TOKENIZER_GET_TEXT_LENGTH)(struct AILIATokenizer *, unsigned int *)
Definition: ailia_speech.h:307
int AILIA_API ailiaSpeechOpenModelFileA(struct AILIASpeech *net, const char *encoder_path, const char *decoder_path, int model_type)
モデルを指定します。(MBSC)
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_TOKENIZER_DECODE)(struct AILIATokenizer *, const int *, unsigned int)
Definition: ailia_speech.h:306
int AILIA_API ailiaSpeechGetText(struct AILIASpeech *net, AILIASpeechText *text, unsigned int version, unsigned int idx)
認識したテキストを取得します。
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_GET_BLOB_INDEX_BY_INPUT_INDEX)(struct AILIANetwork *, unsigned int *, unsigned int)
Definition: ailia_speech.h:321
int AILIA_API ailiaSpeechSetIntermediateCallback(struct AILIASpeech *net, AILIA_SPEECH_USER_API_INTERMEDIATE_CALLBACK callback, void *handle)
認識の途中結果を取得するコールバックを設定します。
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_TOKENIZER_UTF32_TO_UTF8)(char *, unsigned int *, unsigned int)
Definition: ailia_speech.h:311
int AILIA_API ailiaSpeechOpenDiarizationFileW(struct AILIASpeech *net, const wchar_t *segmentation_path, const wchar_t *embedding_path, int type)
話者分離に適用するAIモデルを指定します。(UTF16)
#define AILIA_SPEECH_USER_API
Definition: ailia_speech.h:290
int AILIA_API ailiaSpeechSetLanguage(struct AILIASpeech *net, const char *language)
言語設定を行います。
int AILIA_API ailiaSpeechPostProcess(struct AILIASpeech *net)
後処理を行います。
struct _AILIASpeechText AILIASpeechText
int AILIA_API ailiaSpeechOpenDictionaryFileW(struct AILIASpeech *net, const wchar_t *dictionary_path, int dictionary_type)
誤り訂正辞書を指定します。(UTF16)
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_COPY_BLOB_DATA)(struct AILIANetwork *dst_net, unsigned int dst_blob_idx, struct AILIANetwork *src_net, unsigned int src_blob_idx)
Definition: ailia_speech.h:328
int AILIA_API ailiaSpeechGetTextCount(struct AILIASpeech *net, unsigned int *count)
認識したテキストの数を取得します。
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_OPEN_WEIGHT_FILE_A)(struct AILIANetwork *, const char *)
Definition: ailia_speech.h:315
void(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_DESTROY)(struct AILIANetwork *)
Definition: ailia_speech.h:319
int AILIA_API ailiaSpeechOpenPostProcessFileA(struct AILIASpeech *net, const char *encoder_path, const char *decoder_path, const char *source_path, const char *target_path, const char *prefix, int post_process_type)
後処理に適用するAIモデルを指定します。(MBSC)
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_AUDIO_GET_MEL_SPECTROGRAM)(void *, const void *, int, int, int, int, int, int, int, int, float, int, float, float, int, int, int)
Definition: ailia_speech.h:295
int AILIA_API ailiaSpeechSetConstraint(struct AILIASpeech *net, const char *constraint, int type)
制約の設定を行います。
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_TOKENIZER_OPEN_MODEL_FILE_W)(struct AILIATokenizer *, const wchar_t *)
Definition: ailia_speech.h:302
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_GET_BLOB_DATA)(struct AILIANetwork *, void *, unsigned int, unsigned int)
Definition: ailia_speech.h:323
int AILIA_API ailiaSpeechResetTranscribeState(struct AILIASpeech *net)
ネットワークオブジェクトの内部状態を初期化します。
int AILIA_API ailiaSpeechOpenModelFileW(struct AILIASpeech *net, const wchar_t *encoder_path, const wchar_t *decoder_path, int model_type)
モデルを指定します。(UTF16)
int AILIA_API ailiaSpeechComplete(struct AILIASpeech *net, unsigned int *complete)
全てのデータの処理が完了したかどうかを判定します。
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_SET_MEMORY_MODE)(struct AILIANetwork *, unsigned int)
Definition: ailia_speech.h:318
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_TOKENIZER_GET_TEXT)(struct AILIATokenizer *, char *, unsigned int)
Definition: ailia_speech.h:308
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_TOKENIZER_UTF8_TO_UTF32)(unsigned int *, unsigned int *, const char *, unsigned int)
Definition: ailia_speech.h:310
int AILIA_API ailiaSpeechFinalizeInputData(struct AILIASpeech *net)
音声のキューへの投入を終了します。
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_TOKENIZER_CREATE)(struct AILIATokenizer **, int, int)
Definition: ailia_speech.h:300
const char *AILIA_API ailiaSpeechGetErrorDetail(struct AILIASpeech *net)
エラーの詳細を返します
int AILIA_API ailiaSpeechCreate(struct AILIASpeech **net, int env_id, int num_thread, int memory_mode, int task, int flags, AILIASpeechApiCallback callback, int version)
ネットワークオブジェクトを作成します。
int AILIA_API ailiaSpeechTranscribe(struct AILIASpeech *net)
音声認識を行います。
int AILIA_API ailiaSpeechOpenPostProcessFileW(struct AILIASpeech *net, const wchar_t *encoder_path, const wchar_t *decoder_path, const wchar_t *source_path, const wchar_t *target_path, const char *prefix, int post_process_type)
後処理に適用するAIモデルを指定します。(UTF16)
int AILIA_API ailiaSpeechSetSilentThreshold(struct AILIASpeech *net, float silent_threshold, float speech_sec, float no_speech_sec)
無音判定の閾値を設定します。
int AILIA_API ailiaSpeechOpenDiarizationFileA(struct AILIASpeech *net, const char *segmentation_path, const char *embedding_path, int type)
話者分離に適用するAIモデルを指定します。(MBSC)
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_GET_BLOB_INDEX_BY_OUTPUT_INDEX)(struct AILIANetwork *, unsigned int *, unsigned int)
Definition: ailia_speech.h:322
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_SET_INPUT_BLOB_SHAPE)(struct AILIANetwork *, const AILIAShape *, unsigned int, unsigned int)
Definition: ailia_speech.h:325
const char *(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_GET_ERROR_DETAIL)(struct AILIANetwork *)
Definition: ailia_speech.h:327
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_TOKENIZER_GET_TOKENS)(struct AILIATokenizer *, int *, unsigned int)
Definition: ailia_speech.h:305
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_UPDATE)(struct AILIANetwork *)
Definition: ailia_speech.h:320
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_OPEN_WEIGHT_FILE_W)(struct AILIANetwork *, const wchar_t *)
Definition: ailia_speech.h:316
int AILIA_API ailiaSpeechOpenDictionaryFileA(struct AILIASpeech *net, const char *dictionary_path, int dictionary_type)
誤り訂正辞書を指定します。(MBSC)
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_GET_ENVIRONMENT)(AILIAEnvironment **env, unsigned int env_idx, unsigned int version)
Definition: ailia_speech.h:329
int AILIA_API ailiaSpeechSetPrompt(struct AILIASpeech *net, const char *prompt)
プロンプトの設定を行います。
int(AILIA_SPEECH_USER_API * AILIA_SPEECH_USER_API_AILIA_OPEN_WEIGHT_MEM)(struct AILIANetwork *, const void *, unsigned int)
Definition: ailia_speech.h:317
Definition: ailia_speech.h:343
AILIA_SPEECH_USER_API_AILIA_AUDIO_RESAMPLE ailiaAudioResample
Definition: ailia_speech.h:346
AILIA_SPEECH_USER_API_AILIA_SET_MEMORY_MODE ailiaSetMemoryMode
Definition: ailia_speech.h:364
AILIA_SPEECH_USER_API_AILIA_TOKENIZER_GET_TOKEN_COUNT ailiaTokenizerGetTokenCount
Definition: ailia_speech.h:352
AILIA_SPEECH_USER_API_AILIA_GET_BLOB_SHAPE ailiaGetBlobShape
Definition: ailia_speech.h:372
AILIA_SPEECH_USER_API_AILIA_TOKENIZER_DESTROY ailiaTokenizerDestroy
Definition: ailia_speech.h:357
AILIA_SPEECH_USER_API_AILIA_TOKENIZER_GET_TEXT_LENGTH ailiaTokenizerGetTextLength
Definition: ailia_speech.h:355
AILIA_SPEECH_USER_API_AILIA_TOKENIZER_UTF32_TO_UTF8 ailiaTokenizerUtf32ToUtf8
Definition: ailia_speech.h:359
AILIA_SPEECH_USER_API_AILIA_TOKENIZER_ENCODE ailiaTokenizerEncode
Definition: ailia_speech.h:351
AILIA_SPEECH_USER_API_AILIA_GET_BLOB_INDEX_BY_INPUT_INDEX ailiaGetBlobIndexByInputIndex
Definition: ailia_speech.h:367
AILIA_SPEECH_USER_API_AILIA_UPDATE ailiaUpdate
Definition: ailia_speech.h:366
AILIA_SPEECH_USER_API_AILIA_GET_BLOB_DATA ailiaGetBlobData
Definition: ailia_speech.h:369
AILIA_SPEECH_USER_API_AILIA_TOKENIZER_GET_TEXT ailiaTokenizerGetText
Definition: ailia_speech.h:356
AILIA_SPEECH_USER_API_AILIA_AUDIO_GET_RESAMPLE_LEN ailiaAudioGetResampleLen
Definition: ailia_speech.h:347
AILIA_SPEECH_USER_API_AILIA_CREATE ailiaCreate
Definition: ailia_speech.h:360
AILIA_SPEECH_USER_API_AILIA_OPEN_WEIGHT_FILE_A ailiaOpenWeightFileA
Definition: ailia_speech.h:361
AILIA_SPEECH_USER_API_AILIA_TOKENIZER_UTF8_TO_UTF32 ailiaTokenizerUtf8ToUtf32
Definition: ailia_speech.h:358
AILIA_SPEECH_USER_API_AILIA_SET_INPUT_BLOB_SHAPE ailiaSetInputBlobShape
Definition: ailia_speech.h:371
AILIA_SPEECH_USER_API_AILIA_SET_INPUT_BLOB_DATA ailiaSetInputBlobData
Definition: ailia_speech.h:370
AILIA_SPEECH_USER_API_AILIA_AUDIO_GET_FRAME_LEN ailiaAudioGetFrameLen
Definition: ailia_speech.h:344
AILIA_SPEECH_USER_API_AILIA_AUDIO_GET_MEL_SPECTROGRAM ailiaAudioGetMelSpectrogram
Definition: ailia_speech.h:345
AILIA_SPEECH_USER_API_AILIA_OPEN_WEIGHT_MEM ailiaOpenWeightMem
Definition: ailia_speech.h:363
AILIA_SPEECH_USER_API_AILIA_TOKENIZER_DECODE ailiaTokenizerDecode
Definition: ailia_speech.h:354
AILIA_SPEECH_USER_API_AILIA_OPEN_WEIGHT_FILE_W ailiaOpenWeightFileW
Definition: ailia_speech.h:362
AILIA_SPEECH_USER_API_AILIA_DESTROY ailiaDestroy
Definition: ailia_speech.h:365
AILIA_SPEECH_USER_API_AILIA_COPY_BLOB_DATA ailiaCopyBlobData
Definition: ailia_speech.h:374
AILIA_SPEECH_USER_API_AILIA_TOKENIZER_GET_TOKENS ailiaTokenizerGetTokens
Definition: ailia_speech.h:353
AILIA_SPEECH_USER_API_AILIA_GET_BLOB_INDEX_BY_OUTPUT_INDEX ailiaGetBlobIndexByOutputIndex
Definition: ailia_speech.h:368
AILIA_SPEECH_USER_API_AILIA_GET_ENVIRONMENT ailiaGetEnvironment
Definition: ailia_speech.h:375
AILIA_SPEECH_USER_API_AILIA_TOKENIZER_OPEN_MODEL_FILE_W ailiaTokenizerOpenModelFileW
Definition: ailia_speech.h:350
AILIA_SPEECH_USER_API_AILIA_TOKENIZER_OPEN_MODEL_FILE_A ailiaTokenizerOpenModelFileA
Definition: ailia_speech.h:349
AILIA_SPEECH_USER_API_AILIA_GET_ERROR_DETAIL ailiaGetErrorDetail
Definition: ailia_speech.h:373
AILIA_SPEECH_USER_API_AILIA_TOKENIZER_CREATE ailiaTokenizerCreate
Definition: ailia_speech.h:348
Definition: ailia_speech.h:427
unsigned int speaker_id
Definition: ailia_speech.h:431
float confidence
Definition: ailia_speech.h:433
float time_stamp_end
Definition: ailia_speech.h:430
const char * language
Definition: ailia_speech.h:432
const char * text
Definition: ailia_speech.h:428
float time_stamp_begin
Definition: ailia_speech.h:429