Configuration for keyword spotting. More...

#include <c-api.h>

Collaboration diagram for SherpaOnnxKeywordSpotterConfig:

Public Attributes
SherpaOnnxFeatureConfig	feat_config

SherpaOnnxOnlineModelConfig	model_config

int32_t	max_active_paths

int32_t	num_trailing_blanks

float	keywords_score

float	keywords_threshold

const char *	keywords_file

const char *	keywords_buf

int32_t	keywords_buf_size

Detailed Description

The acoustic model is configured through model_config. In practice this is usually a streaming transducer model.

Keyword definitions can be provided either through keywords_file or through keywords_buf/keywords_buf_size. If both are set, the buffer is used.

Example using sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01-mobile:

SherpaOnnxKeywordSpotterConfig config;
memset(&config, 0, sizeof(config));
 
config.model_config.transducer.encoder =
    "./sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01-mobile/"
    "encoder-epoch-12-avg-2-chunk-16-left-64.int8.onnx";
config.model_config.transducer.decoder =
    "./sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01-mobile/"
    "decoder-epoch-12-avg-2-chunk-16-left-64.onnx";
config.model_config.transducer.joiner =
    "./sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01-mobile/"
    "joiner-epoch-12-avg-2-chunk-16-left-64.int8.onnx";
config.model_config.tokens =
    "./sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01-mobile/"
    "tokens.txt";
config.model_config.provider = "cpu";
config.model_config.num_threads = 1;
 
config.keywords_file =
    "./sherpa-onnx-kws-zipformer-wenetspeech-3.3M-2024-01-01-mobile/"
    "test_wavs/test_keywords.txt";
config.max_active_paths = 4;
config.keywords_score = 3.0f;
config.keywords_threshold = 0.1f;

Definition at line 1662 of file c-api.h.

Member Data Documentation

◆ feat_config

SherpaOnnxFeatureConfig SherpaOnnxKeywordSpotterConfig::feat_config

Feature extraction parameters.

Definition at line 1664 of file c-api.h.

◆ keywords_buf

const char* SherpaOnnxKeywordSpotterConfig::keywords_buf

Optional in-memory keyword data. If non-null, it overrides keywords_file.

Definition at line 1679 of file c-api.h.

◆ keywords_buf_size

int32_t SherpaOnnxKeywordSpotterConfig::keywords_buf_size

Size in bytes of keywords_buf, excluding any trailing ‘’\0'`.

Definition at line 1681 of file c-api.h.

◆ keywords_file

const char* SherpaOnnxKeywordSpotterConfig::keywords_file

Optional keyword file.

Definition at line 1676 of file c-api.h.

◆ keywords_score

float SherpaOnnxKeywordSpotterConfig::keywords_score

Bonus score applied to keywords during search.

Definition at line 1672 of file c-api.h.

◆ keywords_threshold

float SherpaOnnxKeywordSpotterConfig::keywords_threshold

Detection threshold. Larger values are more conservative.

Definition at line 1674 of file c-api.h.

◆ max_active_paths

int32_t SherpaOnnxKeywordSpotterConfig::max_active_paths

Maximum number of active decoding paths.

Definition at line 1668 of file c-api.h.

◆ model_config

SherpaOnnxOnlineModelConfig SherpaOnnxKeywordSpotterConfig::model_config

Streaming acoustic model configuration.

Definition at line 1666 of file c-api.h.

◆ num_trailing_blanks

int32_t SherpaOnnxKeywordSpotterConfig::num_trailing_blanks

Number of trailing blank symbols required before trigger finalization.

Definition at line 1670 of file c-api.h.

The documentation for this struct was generated from the following file:

c-api/c-api.h

Public Attributes