wanson asr development guide
1、Functional Overview
wanson asr Library is mainly used for offline speech recognition, and can support custom wake-up words and command words through model training.
2、development data
The armino platform has adapted and integrated wanson’s offline speech recognition library as a platform component, and provided a project for Huazhen’s speech recognition application development. Customers can develop based on this project.
For a detailed description of the API interface of the wanson-asr library, please refer to the source file:
.\components\bk_thirdparty\asr\wanson\include\asr.hwanson—asr storage path：
.\components\bk_thirdparty\asr\wanson\lib\libasr.a
Note
The wanson Speech Recognition Library requires the audio stream format to be: mono, 16K sampling rate, and 16bit bit width.
wanson Speech Recognition Library is based on floating-point arithmetic.
After modifying the wake word or command word, the
libasr.alibrary needs to be replaced.
3、Real-time offline speech recognition application development
The process of developing real-time offline recognition based on the wanson speech recognition library is as follows:
Initialize speech recognition
Initialize audio sampling
Run speech recognition
Turn on audio sampling
Examples of interface calls are as follows:
/* init wanson asr lib */
Wanson_ASR_Init()
//reset wanson asr
Wanson_ASR_Reset();
/* init mic record */
aud_intf_drv_setup.aud_intf_tx_mic_data = aud_asr_handle;
//init audio component
bk_aud_intf_drv_init(&aud_intf_drv_setup);
aud_work_mode = AUD_INTF_WORK_MODE_GENERAL;
//set audio component work mode
bk_aud_intf_set_mode(aud_work_mode);
//init audio mic
aud_intf_mic_setup.samp_rate = AUD_ADC_SAMP_RATE_16K;
ret = bk_aud_intf_mic_init(&aud_intf_mic_setup);
/* start Speech Recognition */
//Continuously send the collected data to the algorithm for recognition
Wanson_ASR_Recog((short*)asr_buff, 480, &text, &score);
/* turn on audio sampling */
bk_aud_intf_mic_start();
Please refer to the source code for the complete demo： ``\components\demos\media\audio\audio_asr``