Crate vosk_sys

Structs

VoskBatchModel

VoskBatchRecognizer

VoskModel

VoskRecognizer

VoskSpkModel

Functions

vosk_batch_model_free^⚠

Releases batch model object

vosk_batch_model_new^⚠

Creates the batch recognizer object

vosk_batch_model_wait^⚠

Wait for the processing

vosk_batch_recognizer_accept_waveform^⚠

Accept batch voice data

vosk_batch_recognizer_finish_stream^⚠

Closes the stream

vosk_batch_recognizer_free^⚠

Releases batch recognizer object

vosk_batch_recognizer_front_result^⚠

Return results

vosk_batch_recognizer_get_pending_chunks^⚠

Get amount of pending chunks for more intelligent waiting

vosk_batch_recognizer_new^⚠

Creates batch recognizer object @returns recognizer object or NULL if problem occured

vosk_batch_recognizer_pop^⚠

Release and free first retrieved result

vosk_batch_recognizer_set_nlsml^⚠

Set NLSML output @param nlsml - boolean value

vosk_gpu_init^⚠

Init, automatically select a CUDA device and allow multithreading. Must be called once from the main thread. Has no effect if HAVE_CUDA flag is not set.

vosk_gpu_thread_init^⚠

Init CUDA device in a multi-threaded environment. Must be called for each thread. Has no effect if HAVE_CUDA flag is not set.

vosk_model_find_word^⚠

Check if a word can be recognized by the model @param word: the word @returns the word symbol if @param word exists inside the model or -1 otherwise. Reminding that word symbol 0 is for <epsilon>

vosk_model_free^⚠

Releases the model memory

vosk_model_new^⚠

Loads model data from the file and returns the model object

vosk_recognizer_accept_waveform^⚠

Accept voice data

vosk_recognizer_accept_waveform_f^⚠

Same as above but the version with the float data for language bindings where you have audio as array of floats

vosk_recognizer_accept_waveform_s^⚠

Same as above but the version with the short data for language bindings where you have audio as array of shorts

vosk_recognizer_final_result^⚠

Returns speech recognition result. Same as result, but doesn’t wait for silence You usually call it in the end of the stream to get final bits of audio. It flushes the feature pipeline, so all remaining audio chunks got processed.

vosk_recognizer_free^⚠

Releases recognizer object

vosk_recognizer_new^⚠

Creates the recognizer object

vosk_recognizer_new_grm^⚠

Creates the recognizer object with the phrase list

vosk_recognizer_new_spk^⚠

Creates the recognizer object with speaker recognition

vosk_recognizer_partial_result^⚠

Returns partial speech recognition

vosk_recognizer_reset^⚠

Resets the recognizer

vosk_recognizer_result^⚠

Returns speech recognition result

vosk_recognizer_set_max_alternatives^⚠

Configures recognizer to output n-best results

vosk_recognizer_set_nlsml^⚠

Set NLSML output @param nlsml - boolean value

vosk_recognizer_set_partial_words^⚠

Like above return words and confidences in partial results

vosk_recognizer_set_spk_model^⚠

Adds speaker model to already initialized recognizer

vosk_recognizer_set_words^⚠

Enables words with times in the output

vosk_set_log_level^⚠

Set log level for Kaldi messages

vosk_spk_model_free^⚠

Releases the model memory

vosk_spk_model_new^⚠

Loads speaker model data from the file and returns the model object