Configuration for a non-streaming ASR recognizer. More...

#include <c-api.h>

Collaboration diagram for SherpaOnnxOfflineRecognizerConfig:

Public Attributes
SherpaOnnxFeatureConfig	feat_config

SherpaOnnxOfflineModelConfig	model_config

SherpaOnnxOfflineLMConfig	lm_config

const char *	decoding_method

int32_t	max_active_paths

const char *	hotwords_file

float	hotwords_score

const char *	rule_fsts

const char *	rule_fars

float	blank_penalty

SherpaOnnxHomophoneReplacerConfig	hr

Detailed Description

Zero-initialize this struct before use.

Example using Whisper:

SherpaOnnxOfflineRecognizerConfig config;
memset(&config, 0, sizeof(config));
 
config.feat_config.sample_rate = 16000;
config.feat_config.feature_dim = 80;
 
config.model_config.whisper.encoder =
    "./sherpa-onnx-whisper-tiny/tiny-encoder.onnx";
config.model_config.whisper.decoder =
    "./sherpa-onnx-whisper-tiny/tiny-decoder.onnx";
config.model_config.whisper.language = "en";
config.model_config.whisper.task = "transcribe";
config.model_config.tokens =
    "./sherpa-onnx-whisper-tiny/tiny-tokens.txt";
config.model_config.provider = "cpu";
config.model_config.num_threads = 1;
 
config.decoding_method = "greedy_search";

Example using SenseVoice:

config.model_config.sense_voice.model =
    "./sherpa-onnx-sense-voice-zh-en-ja-ko-yue-2024-07-17-int8/model.int8.onnx";
config.model_config.sense_voice.language = "auto";
config.model_config.sense_voice.use_itn = 1;
config.model_config.tokens =
    "./sherpa-onnx-sense-voice-zh-en-ja-ko-yue-2024-07-17-int8/tokens.txt";

Example using Parakeet TDT:

config.model_config.transducer.encoder =
    "./sherpa-onnx-nemo-parakeet-tdt-0.6b-v3-int8/encoder.int8.onnx";
config.model_config.transducer.decoder =
    "./sherpa-onnx-nemo-parakeet-tdt-0.6b-v3-int8/decoder.int8.onnx";
config.model_config.transducer.joiner =
    "./sherpa-onnx-nemo-parakeet-tdt-0.6b-v3-int8/joiner.int8.onnx";
config.model_config.tokens =
    "./sherpa-onnx-nemo-parakeet-tdt-0.6b-v3-int8/tokens.txt";
config.model_config.model_type = "nemo_transducer";

Definition at line 1155 of file c-api.h.

Member Data Documentation

◆ blank_penalty

float SherpaOnnxOfflineRecognizerConfig::blank_penalty

Optional blank penalty applied during decoding.

Definition at line 1178 of file c-api.h.

◆ decoding_method

const char* SherpaOnnxOfflineRecognizerConfig::decoding_method

Decoding method, for example "greedy_search" or "modified_beam_search".

Definition at line 1164 of file c-api.h.

◆ feat_config

SherpaOnnxFeatureConfig SherpaOnnxOfflineRecognizerConfig::feat_config

Feature extraction settings.

Definition at line 1157 of file c-api.h.

◆ hotwords_file

const char* SherpaOnnxOfflineRecognizerConfig::hotwords_file

Path to a hotwords file.

Definition at line 1169 of file c-api.h.

◆ hotwords_score

float SherpaOnnxOfflineRecognizerConfig::hotwords_score

Bonus score added to each hotword token.

Definition at line 1172 of file c-api.h.

◆ hr

SherpaOnnxHomophoneReplacerConfig SherpaOnnxOfflineRecognizerConfig::hr

Optional homophone replacement configuration.

Definition at line 1181 of file c-api.h.

◆ lm_config

SherpaOnnxOfflineLMConfig SherpaOnnxOfflineRecognizerConfig::lm_config

Optional language model configuration.

Definition at line 1161 of file c-api.h.

◆ max_active_paths

int32_t SherpaOnnxOfflineRecognizerConfig::max_active_paths

Number of active paths for modified beam search.

Definition at line 1166 of file c-api.h.

◆ model_config

SherpaOnnxOfflineModelConfig SherpaOnnxOfflineRecognizerConfig::model_config

Offline model configuration.

Definition at line 1159 of file c-api.h.

◆ rule_fars

const char* SherpaOnnxOfflineRecognizerConfig::rule_fars

Path to FAR archives used by text-processing rules.

Definition at line 1176 of file c-api.h.

◆ rule_fsts

const char* SherpaOnnxOfflineRecognizerConfig::rule_fsts

Path to punctuation or text-processing rule FSTs.

Definition at line 1174 of file c-api.h.

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

c-api/c-api.h

Public Attributes