Sherpa ONNX

https://docs.rs/sherpa-onnx/latest/sherpa_onnx/
Admin
Sherpa ONNX is a Rust library for speech recognition and text-to-speech using ONNX models,...
Tokens:16,124
Snippets:196
Trust Score:9.7
Update:2 months ago
Show doc for...
Tokens:
Raw
### OfflineRecognizer Usage Example

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineRecognizer.html

Demonstrates how to configure and use an OfflineRecognizer to decode audio from a wave file.

```rust
use sherpa_onnx::{
    OfflineRecognizer, OfflineRecognizerConfig, OfflineTransducerModelConfig, Wave,
};

let wave = Wave::read("./test.wav").expect("read wave");
let mut config = OfflineRecognizerConfig::default();
config.model_config.transducer = OfflineTransducerModelConfig {
    encoder: Some("./sherpa-onnx-nemo-parakeet-tdt-0.6b-v3-int8/encoder.int8.onnx".into()),
    decoder: Some("./sherpa-onnx-nemo-parakeet-tdt-0.6b-v3-int8/decoder.int8.onnx".into()),
    joiner: Some("./sherpa-onnx-nemo-parakeet-tdt-0.6b-v3-int8/joiner.int8.onnx".into()),
};
config.model_config.tokens =
    Some("./sherpa-onnx-nemo-parakeet-tdt-0.6b-v3-int8/tokens.txt".into());
config.model_config.model_type = Some("nemo_transducer".into());

let recognizer = OfflineRecognizer::create(&config).expect("create recognizer");
let stream = recognizer.create_stream();
stream.accept_waveform(wave.sample_rate(), wave.samples());
recognizer.decode(&stream);
let result = stream.get_result().expect("result");
println!("{}", result.text);
```

--------------------------------

### Get sherpa-onnx version

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/fn.version.html

Returns the version string compiled into the native library.

```rust
pub fn version() -> &'static str
```

--------------------------------

### OfflineRecognizerConfig Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineRecognizerConfig.html

The primary configuration structure for offline recognition. Use the Default implementation as a starting point before customizing fields.

```rust
pub struct OfflineRecognizerConfig {
    pub feat_config: FeatureConfig,
    pub model_config: OfflineModelConfig,
    pub lm_config: OfflineLMConfig,
    pub decoding_method: Option<String>,
    pub max_active_paths: i32,
    pub hotwords_file: Option<String>,
    pub hotwords_score: f32,
    pub rule_fsts: Option<String>,
    pub rule_fars: Option<String>,
    pub blank_penalty: f32,
    pub hr: HomophoneReplacerConfig,
}
```

--------------------------------

### KeywordSpotter::create

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.KeywordSpotter.html

Creates a new instance of the KeywordSpotter using the provided configuration.

```APIDOC
## pub fn create(config: &KeywordSpotterConfig) -> Option<Self>

### Description
Creates a keyword spotter instance from a given KeywordSpotterConfig.

### Parameters
- **config** (&KeywordSpotterConfig) - Required - The configuration object for the keyword spotter.
```

--------------------------------

### AudioTagging::create

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.AudioTagging.html

Creates a new instance of the AudioTagging struct using the provided configuration.

```APIDOC
## pub fn create(config: &AudioTaggingConfig) -> Option<Self>

### Description
Creates a tagger from the provided configuration.

### Parameters
- **config** (&AudioTaggingConfig) - Required - The configuration object for the audio tagger.

### Returns
- **Option<Self>** - Returns an instance of AudioTagging if successful, otherwise None.
```

--------------------------------

### pub fn samples(&self) -> &[f32]

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.GeneratedAudio.html

Retrieves the generated audio samples as a slice of 32-bit floating-point numbers.

```APIDOC
## pub fn samples(&self) -> &[f32]

### Description
Borrow the generated audio samples as a slice of f32 values.

### Returns
- **&[f32]** - A slice containing the generated audio samples.
```

--------------------------------

### DisplayManager::new

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.DisplayManager.html

Creates a new, empty instance of the DisplayManager.

```APIDOC
## pub fn new() -> Self

### Description
Creates an empty display manager instance.

### Signature
`pub fn new() -> Self`
```

--------------------------------

### Initialize OfflineTts Engine

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineTts.html

Demonstrates how to configure and initialize an OfflineTts engine using model configuration paths.

```rust
use sherpa_onnx::{
    OfflineTts, OfflineTtsConfig, OfflineTtsModelConfig, OfflineTtsPocketModelConfig,
};

let config = OfflineTtsConfig {
    model: OfflineTtsModelConfig {
        pocket: OfflineTtsPocketModelConfig {
            lm_flow: Some("./sherpa-onnx-pocket-tts-int8-2026-01-26/lm_flow.int8.onnx".into()),
            lm_main: Some("./sherpa-onnx-pocket-tts-int8-2026-01-26/lm_main.int8.onnx".into()),
            encoder: Some("./sherpa-onnx-pocket-tts-int8-2026-01-26/encoder.onnx".into()),
            decoder: Some("./sherpa-onnx-pocket-tts-int8-2026-01-26/decoder.int8.onnx".into()),
            text_conditioner: Some(
                "./sherpa-onnx-pocket-tts-int8-2026-01-26/text_conditioner.onnx".into(),
            ),
            vocab_json: Some("./sherpa-onnx-pocket-tts-int8-2026-01-26/vocab.json".into()),
            token_scores_json: Some(
                "./sherpa-onnx-pocket-tts-int8-2026-01-26/token_scores.json".into(),
            ),
            ..Default::default()
        },
        ..Default::default()
    },
    ..Default::default()
};

let tts = OfflineTts::create(&config).expect("create tts");
println!("{}", tts.sample_rate());
```

--------------------------------

### pub fn version() -> &'static str

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/fn.version.html

Returns the sherpa-onnx version string compiled into the native library.

```APIDOC
## Function: version

### Description
Returns the sherpa-onnx version string compiled into the native library.

### Signature
`pub fn version() -> &'static str`
```

--------------------------------

### pub fn write(filename: &str, samples: &[f32], sample_rate: i32) -> bool

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/fn.write.html

Writes normalized PCM samples to a WAV file. This is a convenient utility when working with raw f32 sample data.

```APIDOC
## Function: write

### Description
Writes normalized PCM samples to a WAV file. This is convenient when an API returns a plain `Vec<f32>` and you do not need to build a `Wave` first.

### Signature
`pub fn write(filename: &str, samples: &[f32], sample_rate: i32) -> bool`

### Parameters
- **filename** (&str) - The path to the output WAV file.
- **samples** (&[f32]) - A slice of normalized PCM samples.
- **sample_rate** (i32) - The sample rate of the audio data.

### Returns
- **bool** - Returns true if the file was written successfully, false otherwise.
```

--------------------------------

### write

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/index.html

Writes PCM samples to a WAV file.

```APIDOC
## write

### Description
Write normalized PCM samples to a WAV file.
```

--------------------------------

### pub fn save(&self, filename: &str) -> bool

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.GeneratedAudio.html

Saves the generated audio data to a specified WAV file path.

```APIDOC
## pub fn save(&self, filename: &str) -> bool

### Description
Saves the generated audio to a WAV file at the provided path.

### Parameters
- **filename** (&str) - Required - The path where the WAV file should be saved.

### Returns
- **bool** - Returns true if the file was saved successfully, false otherwise.
```

--------------------------------

### Add sherpa-onnx dependency

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/index.html

The default configuration for most users, which links the library statically.

```toml
sherpa-onnx = "1.13.4"
```

--------------------------------

### KeywordSpotter::create_stream

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.KeywordSpotter.html

Creates a stream for the keyword spotter using the default configured keywords.

```APIDOC
## pub fn create_stream(&self) -> OnlineStream

### Description
Creates a stream that uses the keywords defined in the KeywordSpotterConfig.
```

--------------------------------

### KeywordSpotter::create_stream_with_keywords

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.KeywordSpotter.html

Creates a stream using a custom list of keywords.

```APIDOC
## pub fn create_stream_with_keywords(&self, keywords: &str) -> OnlineStream

### Description
Creates a stream that uses the provided keywords string instead of the default configuration.

### Parameters
- **keywords** (&str) - Required - A string containing the keywords to use.
```

--------------------------------

### pub fn file_exists(filename: &str) -> bool

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/fn.file_exists.html

Checks if a file exists at the specified path.

```APIDOC
## file_exists

### Description
Returns true if the specified filename exists according to the native helper.

### Signature
`pub fn file_exists(filename: &str) -> bool`

### Parameters
- **filename** (&str) - The path to the file to check.

### Returns
- **bool** - Returns true if the file exists, false otherwise.
```

--------------------------------

### OnlinePunctuation::create

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlinePunctuation.html

Creates a new instance of the OnlinePunctuation restorer using the provided configuration.

```APIDOC
## pub fn create(config: &OnlinePunctuationConfig) -> Option<Self>

### Description
Creates a new OnlinePunctuation instance based on the provided configuration.

### Parameters
- **config** (&OnlinePunctuationConfig) - Required - The configuration settings for the punctuator.

### Returns
- **Option<Self>** - Returns an instance of OnlinePunctuation if successful, otherwise None.
```

--------------------------------

### VoiceActivityDetector::create

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.VoiceActivityDetector.html

Creates a new instance of the VoiceActivityDetector with the specified configuration and buffer size.

```APIDOC
## pub fn create(config: &VadModelConfig, buffer_size_in_seconds: f32) -> Option<Self>

### Description
Initializes a new voice activity detector using the provided model configuration and buffer size.

### Parameters
- **config** (&VadModelConfig) - Required - The configuration for the VAD model.
- **buffer_size_in_seconds** (f32) - Required - The size of the internal result buffer in seconds.
```

--------------------------------

### Streaming speech recognition with OnlineRecognizer

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/index.html

Configures and runs a streaming speech recognizer using a Zipformer model. Requires a valid wave file and model configuration paths.

```rust
use sherpa_onnx::{OnlineRecognizer, OnlineRecognizerConfig, Wave};

let wave = Wave::read("./test.wav").expect("read wave");

let mut config = OnlineRecognizerConfig::default();
config.model_config.transducer.encoder = Some(
    "./sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20/encoder-epoch-99-avg-1.int8.onnx".into(),
);
config.model_config.transducer.decoder = Some(
    "./sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20/decoder-epoch-99-avg-1.onnx".into(),
);
config.model_config.transducer.joiner = Some(
    "./sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20/joiner-epoch-99-avg-1.int8.onnx".into(),
);
config.model_config.tokens = Some(
    "./sherpa-onnx-streaming-zipformer-bilingual-zh-en-2023-02-20/tokens.txt".into(),
);
config.enable_endpoint = true;
config.decoding_method = Some("greedy_search".into());

let recognizer = OnlineRecognizer::create(&config).expect("create recognizer");
let stream = recognizer.create_stream();
stream.accept_waveform(wave.sample_rate(), wave.samples());
stream.input_finished();
while recognizer.is_ready(&stream) {
    recognizer.decode(&stream);
}
```

--------------------------------

### OnlineSpeechDenoiser::create

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlineSpeechDenoiser.html

Creates a new instance of the OnlineSpeechDenoiser using the provided configuration.

```APIDOC
## pub fn create(config: &OnlineSpeechDenoiserConfig) -> Option<Self>

### Description
Creates a new streaming speech denoiser instance based on the provided configuration object.

### Parameters
- **config** (&OnlineSpeechDenoiserConfig) - Required - The configuration settings for the denoiser.

### Returns
- **Option<Self>** - Returns an instance of the denoiser if initialization is successful, otherwise None.
```

--------------------------------

### OnlineRecognizer::create

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlineRecognizer.html

Creates a new instance of the OnlineRecognizer using the provided configuration.

```APIDOC
## pub fn create(config: &OnlineRecognizerConfig) -> Option<Self>

### Description
Creates a recognizer from the provided configuration.

### Parameters
- **config** (&OnlineRecognizerConfig) - Required - The configuration object for the recognizer.
```

--------------------------------

### version

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/index.html

Retrieves the version string of the native library.

```APIDOC
## version

### Description
Return the sherpa-onnx version string compiled into the native library.
```

--------------------------------

### Set library directory environment variable

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/index.html

Override the default build behavior by pointing to a directory containing pre-existing sherpa-onnx libraries.

```bash
export SHERPA_ONNX_LIB_DIR=/path/to/sherpa-onnx/lib
```

--------------------------------

### get_option

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlineStream.html

Retrieves the value of a configuration option.

```APIDOC
### pub fn get_option(&self, key: &str) -> String

Returns the value associated with the provided key.
```

--------------------------------

### Configure shared library dependency

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/index.html

Use this configuration to disable default static linking and enable shared library support.

```toml
sherpa-onnx = { version = "1.13.4", default-features = false, features = ["shared"] }
```

--------------------------------

### KeywordSpotter struct definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.KeywordSpotter.html

The primary structure for streaming keyword spotting.

```rust
pub struct KeywordSpotter { /* private fields */ }
```

--------------------------------

### Retrieve Git SHA1

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/fn.git_sha1.html

Returns the Git SHA1 of the native library build as a static string.

```rust
pub fn git_sha1() -> &'static str
```

--------------------------------

### Define KeywordSpotterConfig struct

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.KeywordSpotterConfig.html

The structure definition for KeywordSpotterConfig, containing configuration for features, models, and keyword detection parameters.

```rust
pub struct KeywordSpotterConfig {
    pub feat_config: FeatureConfig,
    pub model_config: OnlineModelConfig,
    pub max_active_paths: i32,
    pub num_trailing_blanks: i32,
    pub keywords_score: f32,
    pub keywords_threshold: f32,
    pub keywords_file: Option<String>,
    pub keywords_buf: Option<String>,
}
```

--------------------------------

### OnlineRecognizer::is_ready

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlineRecognizer.html

Checks if the recognizer has sufficient audio data to perform another decoding step.

```APIDOC
## pub fn is_ready(&self, stream: &OnlineStream) -> bool

### Description
Return true if the recognizer has enough audio to run another step.

### Parameters
- **stream** (&OnlineStream) - Required - The stream to check.
```

--------------------------------

### Write PCM samples to WAV file

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/fn.write.html

Use this function to save normalized PCM samples directly to a file. It is useful when working with raw Vec<f32> data.

```rust
pub fn write(filename: &str, samples: &[f32], sample_rate: i32) -> bool
```

--------------------------------

### Offline text-to-speech with OfflineTts

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/index.html

Initializes an offline TTS engine using pocket model configurations. Accesses the sample rate after successful creation.

```rust
use sherpa_onnx::{OfflineTts, OfflineTtsConfig, OfflineTtsModelConfig, OfflineTtsPocketModelConfig};

let config = OfflineTtsConfig {
    model: OfflineTtsModelConfig {
        pocket: OfflineTtsPocketModelConfig {
            lm_flow: Some("./sherpa-onnx-pocket-tts-int8-2026-01-26/lm_flow.int8.onnx".into()),
            lm_main: Some("./sherpa-onnx-pocket-tts-int8-2026-01-26/lm_main.int8.onnx".into()),
            encoder: Some("./sherpa-onnx-pocket-tts-int8-2026-01-26/encoder.onnx".into()),
            decoder: Some("./sherpa-onnx-pocket-tts-int8-2026-01-26/decoder.int8.onnx".into()),
            text_conditioner: Some(
                "./sherpa-onnx-pocket-tts-int8-2026-01-26/text_conditioner.onnx".into(),
            ),
            vocab_json: Some("./sherpa-onnx-pocket-tts-int8-2026-01-26/vocab.json".into()),
            token_scores_json: Some(
                "./sherpa-onnx-pocket-tts-int8-2026-01-26/token_scores.json".into(),
            ),
            ..Default::default()
        },
        ..Default::default()
    },
    ..Default::default()
};

let tts = OfflineTts::create(&config).expect("create tts");
println!("{}", tts.sample_rate());
```

--------------------------------

### git_sha1()

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/fn.git_sha1.html

Returns the Git SHA1 of the native library build as a static string.

```APIDOC
## pub fn git_sha1() -> &'static str

### Description
Returns the Git SHA1 of the native library build.

### Signature
`pub fn git_sha1() -> &'static str`

### Returns
- **&'static str** - The Git SHA1 hash of the native library build.
```

--------------------------------

### Perform offline speech recognition in Rust

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/index.html

Configures an OfflineRecognizer with SenseVoice settings and processes a wave file to extract text. Requires the sherpa_onnx crate and appropriate model files.

```rust
use sherpa_onnx::{
    OfflineRecognizer, OfflineRecognizerConfig, OfflineSenseVoiceModelConfig, Wave,
};

let wave = Wave::read("./test.wav").expect("read wave");

let mut config = OfflineRecognizerConfig::default();
config.model_config.sense_voice = OfflineSenseVoiceModelConfig {
    model: Some(
        "./sherpa-onnx-sense-voice-zh-en-ja-ko-yue-2024-07-17-int8/model.int8.onnx".into(),
    ),
    language: Some("auto".into()),
    use_itn: true,
};
config.model_config.tokens = Some(
    "./sherpa-onnx-sense-voice-zh-en-ja-ko-yue-2024-07-17-int8/tokens.txt".into(),
);

let recognizer = OfflineRecognizer::create(&config).expect("create recognizer");
let stream = recognizer.create_stream();
stream.accept_waveform(wave.sample_rate(), wave.samples());
recognizer.decode(&stream);

let result = stream.get_result().expect("result");
println!("{}", result.text);
```

--------------------------------

### LinearResampler::create

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.LinearResampler.html

Creates a new instance of LinearResampler configured for specific input and output sample rates.

```APIDOC
## pub fn create(samp_rate_in_hz: i32, samp_rate_out_hz: i32) -> Option<Self>

### Description
Creates a new resampler that converts audio from the specified input sample rate to the output sample rate.

### Parameters
- **samp_rate_in_hz** (i32) - The input sample rate in Hz.
- **samp_rate_out_hz** (i32) - The output sample rate in Hz.

### Returns
- **Option<Self>** - Returns an instance of LinearResampler if successful, otherwise None.
```

--------------------------------

### Retrieve Git date of native library

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/fn.git_date.html

Returns the Git date of the native library build as a static string.

```rust
pub fn git_date() -> &'static str
```

--------------------------------

### OnlineRecognizer::create_stream

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlineRecognizer.html

Initializes an empty online stream for audio processing.

```APIDOC
## pub fn create_stream(&self) -> OnlineStream

### Description
Creates an empty online stream.
```

--------------------------------

### pub fn sample_rate(&self) -> i32

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.GeneratedAudio.html

Retrieves the sample rate of the generated audio in Hz.

```APIDOC
## pub fn sample_rate(&self) -> i32

### Description
Returns the output sample rate of the generated audio in Hz.

### Returns
- **i32** - The sample rate in Hz.
```

--------------------------------

### OfflineSpeechDenoiser::run

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineSpeechDenoiser.html

Processes audio samples to perform denoising.

```APIDOC
## pub fn run(&self, samples: &[f32], sample_rate: i32) -> DenoisedAudio

### Description
Denoises a chunk or a complete waveform of audio samples.

### Parameters
- **samples** (&[f32]) - Required - The audio waveform samples to be denoised.
- **sample_rate** (i32) - Required - The sample rate of the input audio.

### Returns
- **DenoisedAudio** - The resulting denoised audio data.
```

--------------------------------

### accept_waveform

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.AudioTaggingOfflineStream.html

Appends waveform samples to the audio tagging stream.

```APIDOC
## pub fn accept_waveform(&self, sample_rate: i32, samples: &[f32])

### Description
Appends waveform samples to the audio clip for offline tagging.

### Parameters
- **sample_rate** (i32) - The sample rate of the audio data.
- **samples** (&[f32]) - A slice containing the audio waveform samples.
```

--------------------------------

### SpeakerEmbeddingExtractor::create

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.SpeakerEmbeddingExtractor.html

Creates a new instance of the SpeakerEmbeddingExtractor using the provided configuration.

```APIDOC
## pub fn create(config: &SpeakerEmbeddingExtractorConfig) -> Option<Self>

### Description
Creates an extractor from the provided configuration.

### Parameters
- **config** (&SpeakerEmbeddingExtractorConfig) - Required - The configuration object for the extractor.

### Returns
- **Option<Self>** - Returns an instance of the extractor if successful, otherwise None.
```

--------------------------------

### CircularBuffer::new

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.CircularBuffer.html

Creates a new instance of CircularBuffer with a specified capacity in samples.

```APIDOC
## pub fn new(capacity: i32) -> Option<Self>

### Description
Creates a new buffer with capacity measured in samples.

### Parameters
- **capacity** (i32) - Required - The capacity of the buffer in samples.

### Returns
- **Option<Self>** - Returns the new CircularBuffer instance if successful, otherwise None.
```

--------------------------------

### git_date()

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/fn.git_date.html

Returns the Git date of the native library build as a static string.

```APIDOC
## git_date

### Description
Returns the Git date of the native library build.

### Signature
`pub fn git_date() -> &'static str`
```

--------------------------------

### Wave Metadata and Data Access

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.Wave.html

Methods to retrieve information about the loaded audio, including sample rate, total sample count, and the raw normalized PCM samples.

```APIDOC
### pub fn sample_rate(&self) -> i32
Return the sample rate in Hz.

### pub fn num_samples(&self) -> i32
Return the number of samples in the waveform.

### pub fn samples(&self) -> &[f32]
Return the normalized PCM samples.
```

--------------------------------

### has_option

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlineStream.html

Checks if a specific configuration option exists.

```APIDOC
### pub fn has_option(&self, key: &str) -> bool

Returns true if the option exists, otherwise false.
```

--------------------------------

### DisplayManager::render

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.DisplayManager.html

Renders the current state of the display manager to stdout.

```APIDOC
## pub fn render(&mut self)

### Description
Render the current state to stdout. Rendering is throttled slightly to reduce terminal flicker.
```

--------------------------------

### AudioTagging::create_stream

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.AudioTagging.html

Creates a stream that accepts one complete audio clip for processing.

```APIDOC
## pub fn create_stream(&self) -> AudioTaggingOfflineStream

### Description
Creates a stream that accepts one complete clip.

### Returns
- **AudioTaggingOfflineStream** - A stream object ready to accept audio data.
```

--------------------------------

### OnlineSpeechDenoiser::run

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlineSpeechDenoiser.html

Processes a chunk of audio samples to perform denoising.

```APIDOC
## pub fn run(&self, samples: &[f32], sample_rate: i32) -> DenoisedAudio

### Description
Processes a single input chunk of audio samples and returns the denoised result.

### Parameters
- **samples** (&[f32]) - Required - A slice of floating-point audio samples.
- **sample_rate** (i32) - Required - The sample rate of the input audio.

### Returns
- **DenoisedAudio** - The resulting denoised audio data.
```

--------------------------------

### set_option

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlineStream.html

Sets a configuration option for the stream.

```APIDOC
### pub fn set_option(&self, key: &str, value: &str)

Sets a specific option for the stream using a key-value pair.
```

--------------------------------

### OnlinePunctuationModelConfig Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlinePunctuationModelConfig.html

The structure definition for configuring online punctuation models.

```rust
pub struct OnlinePunctuationModelConfig {
    pub cnn_bilstm: Option<String>,
    pub bpe_vocab: Option<String>,
    pub num_threads: i32,
    pub debug: bool,
    pub provider: Option<String>,
}
```

--------------------------------

### KeywordSpotter::decode

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.KeywordSpotter.html

Performs an incremental decoding step for a specific stream.

```APIDOC
## pub fn decode(&self, stream: &OnlineStream)

### Description
Decodes one incremental step for the provided stream.

### Parameters
- **stream** (&OnlineStream) - Required - The stream to decode.
```

--------------------------------

### OfflineQwen3ASRModelConfig Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineQwen3ASRModelConfig.html

The structure definition for configuring offline Qwen3 ASR models, containing paths to model components and generation hyperparameters.

```rust
pub struct OfflineQwen3ASRModelConfig {
    pub conv_frontend: Option<String>,
    pub encoder: Option<String>,
    pub decoder: Option<String>,
    pub tokenizer: Option<String>,
    pub max_total_len: i32,
    pub max_new_tokens: i32,
    pub temperature: f32,
    pub top_p: f32,
    pub seed: i32,
    pub hotwords: Option<String>,
}
```

--------------------------------

### OfflineTts::generate_with_config

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineTts.html

Generates audio from the provided text using the specified generation configuration and an optional callback.

```APIDOC
## pub fn generate_with_config<F>( &self, text: &str, config: &GenerationConfig, callback: Option<F>, ) -> Option<GeneratedAudio>

### Description
Generates audio for the given text. The optional callback receives generated samples and a progress value (0 to 1); returning false from the callback stops generation early.

### Parameters
- **text** (&str) - Required - The input text to synthesize.
- **config** (&GenerationConfig) - Required - Configuration for the generation process.
- **callback** (Option<F>) - Optional - A closure that receives generated samples and progress.
```

--------------------------------

### file_exists function signature

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/fn.file_exists.html

The function signature for checking file existence.

```rust
pub fn file_exists(filename: &str) -> bool
```

--------------------------------

### Wave::read

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.Wave.html

Reads a mono WAV file from the specified path. Returns an Option containing the Wave instance, or None if the file cannot be opened or decoded.

```APIDOC
### pub fn read(filename: &str) -> Option<Self>

Reads a mono WAV file from disk. Returns `None` if the file cannot be opened or decoded.
```

--------------------------------

### AudioTaggingModelConfig Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.AudioTaggingModelConfig.html

The structure definition for configuring audio tagging models.

```rust
pub struct AudioTaggingModelConfig {
    pub zipformer: OfflineZipformerAudioTaggingModelConfig,
    pub ced: Option<String>,
    pub num_threads: i32,
    pub debug: bool,
    pub provider: Option<String>,
}
```

--------------------------------

### OfflineSpeechDenoiser::create

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineSpeechDenoiser.html

Creates a new instance of the OfflineSpeechDenoiser using the provided configuration.

```APIDOC
## pub fn create(config: &OfflineSpeechDenoiserConfig) -> Option<Self>

### Description
Creates a denoiser instance from the provided configuration object.

### Parameters
- **config** (&OfflineSpeechDenoiserConfig) - Required - The configuration settings for the denoiser.

### Returns
- **Option<Self>** - Returns an instance of OfflineSpeechDenoiser if successful, otherwise None.
```

--------------------------------

### OnlineRecognizer::create_stream_with_hotwords

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlineRecognizer.html

Initializes an online stream with specific hotwords for improved recognition.

```APIDOC
## pub fn create_stream_with_hotwords(&self, hotwords: &str) -> OnlineStream

### Description
Creates a stream with per-stream hotwords.

### Parameters
- **hotwords** (&str) - Required - The hotwords to be used for this stream.
```

--------------------------------

### KeywordSpotter::get_result

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.KeywordSpotter.html

Retrieves the structured result for a given stream.

```APIDOC
## pub fn get_result(&self, stream: &OnlineStream) -> Option<KeywordResult>

### Description
Returns the structured keyword spotting result for the specified stream.

### Parameters
- **stream** (&OnlineStream) - Required - The stream to retrieve results from.
```

--------------------------------

### DisplayManager::current_text

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.DisplayManager.html

Retrieves the current partial text.

```APIDOC
## pub fn current_text(&self) -> &str

### Description
Borrow the current partial text.
```

--------------------------------

### OfflineTts::create

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineTts.html

Creates a new instance of the OfflineTts engine using the provided configuration.

```APIDOC
## pub fn create(config: &OfflineTtsConfig) -> Option<Self>

### Description
Creates a TTS engine from the provided configuration.

### Parameters
- **config** (&OfflineTtsConfig) - Required - The configuration object containing model paths and settings.
```

--------------------------------

### file_exists

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/index.html

Checks if a file exists using the native helper.

```APIDOC
## file_exists

### Description
Return `true` if `filename` exists according to the native helper.
```

--------------------------------

### Wave::write

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.Wave.html

Writes the current waveform data to a WAV file at the specified path.

```APIDOC
### pub fn write(&self, filename: &str) -> bool

Write this waveform to a WAV file. Returns `true` on success, `false` otherwise.
```

--------------------------------

### Append Waveform Samples

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.AudioTaggingOfflineStream.html

Appends audio waveform samples to the stream for processing.

```rust
pub fn accept_waveform(&self, sample_rate: i32, samples: &[f32])
```

--------------------------------

### SpeakerEmbeddingExtractor::is_ready

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.SpeakerEmbeddingExtractor.html

Checks if the provided stream has accumulated enough audio data to compute an embedding.

```APIDOC
## pub fn is_ready(&self, stream: &OnlineStream) -> bool

### Description
Return true if enough audio has been accumulated to compute an embedding.

### Parameters
- **stream** (&OnlineStream) - Required - The stream to check.

### Returns
- **bool** - True if ready, false otherwise.
```

--------------------------------

### AudioTagging Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.AudioTagging.html

The definition of the AudioTagging struct.

```rust
pub struct AudioTagging { /* private fields */ }
```

--------------------------------

### Define GenerationConfig struct

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.GenerationConfig.html

The structure definition for per-request generation options in sherpa-onnx.

```rust
pub struct GenerationConfig {
    pub silence_scale: f32,
    pub speed: f32,
    pub sid: i32,
    pub reference_audio: Option<Vec<f32>>,
    pub reference_sample_rate: i32,
    pub reference_text: Option<String>,
    pub num_steps: i32,
    pub extra: Option<HashMap<String, Value>>,
}
```

--------------------------------

### Define SileroVadModelConfig Struct

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.SileroVadModelConfig.html

The structure definition for configuring the Silero VAD model.

```rust
pub struct SileroVadModelConfig {
    pub model: Option<String>,
    pub threshold: f32,
    pub min_silence_duration: f32,
    pub min_speech_duration: f32,
    pub window_size: i32,
    pub max_speech_duration: f32,
}
```

--------------------------------

### OnlineSpeechDenoiser::sample_rate

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlineSpeechDenoiser.html

Retrieves the expected sample rate for the denoiser.

```APIDOC
## pub fn sample_rate(&self) -> i32

### Description
Returns the model sample rate that is expected by this denoiser instance.

### Returns
- **i32** - The expected sample rate.
```

--------------------------------

### CircularBuffer::get

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.CircularBuffer.html

Retrieves a specific range of samples from the buffer.

```APIDOC
## pub fn get(&self, start_index: i32, n: i32) -> Vec<f32>

### Description
Copy `n` samples starting at `start_index`.

### Parameters
- **start_index** (i32) - Required - The starting index from which to copy samples.
- **n** (i32) - Required - The number of samples to copy.

### Returns
- **Vec<f32>** - A vector containing the requested samples.
```

--------------------------------

### AudioTagging::compute

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.AudioTagging.html

Computes the top K audio events for the provided stream.

```APIDOC
## pub fn compute(&self, stream: &AudioTaggingOfflineStream, top_k: i32) -> Vec<AudioEvent>

### Description
Compute the top `top_k` events for the provided stream.

### Parameters
- **stream** (&AudioTaggingOfflineStream) - Required - The stream containing the audio data to process.
- **top_k** (i32) - Required - The number of top events to return.

### Returns
- **Vec<AudioEvent>** - A list of the top audio events detected.
```

--------------------------------

### Define OfflineTtsZipvoiceModelConfig struct

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineTtsZipvoiceModelConfig.html

The structure definition for configuring ZipVoice models in sherpa-onnx.

```rust
pub struct OfflineTtsZipvoiceModelConfig {
    pub tokens: Option<String>,
    pub encoder: Option<String>,
    pub decoder: Option<String>,
    pub vocoder: Option<String>,
    pub data_dir: Option<String>,
    pub lexicon: Option<String>,
    pub feat_scale: f32,
    pub t_shift: f32,
    pub target_rms: f32,
    pub guidance_scale: f32,
}
```

--------------------------------

### LinearResampler::resample

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.LinearResampler.html

Resamples a buffer of audio samples.

```APIDOC
## pub fn resample(&self, samples: &[f32], flush: bool) -> Vec<f32>

### Description
Resamples the provided audio samples. Set the flush parameter to true when processing the final chunk of audio to clear internal buffers.

### Parameters
- **samples** (&[f32]) - A slice of audio samples to be resampled.
- **flush** (bool) - Whether to flush the internal buffer after processing the samples.

### Returns
- **Vec<f32>** - A vector containing the resampled audio data.
```

--------------------------------

### CircularBuffer::push

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.CircularBuffer.html

Appends a slice of samples to the tail of the buffer.

```APIDOC
## pub fn push(&self, samples: &[f32])

### Description
Append samples to the tail of the buffer.

### Parameters
- **samples** (&[f32]) - Required - A slice of floating-point samples to add to the buffer.
```

--------------------------------

### OnlineRecognizerConfig Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlineRecognizerConfig.html

The structure definition for OnlineRecognizerConfig, outlining the various configuration fields required for initializing an online recognizer.

```rust
pub struct OnlineRecognizerConfig {Show 16 fields
    pub feat_config: FeatureConfig,
    pub model_config: OnlineModelConfig,
    pub decoding_method: Option<String>,
    pub max_active_paths: i32,
    pub enable_endpoint: bool,
    pub rule1_min_trailing_silence: f32,
    pub rule2_min_trailing_silence: f32,
    pub rule3_min_utterance_length: f32,
    pub hotwords_file: Option<String>,
    pub hotwords_score: f32,
    pub ctc_fst_decoder_config: OnlineCtcFstDecoderConfig,
    pub rule_fsts: Option<String>,
    pub rule_fars: Option<String>,
    pub blank_penalty: f32,
    pub hotwords_buf: Option<Vec<u8>>,
    pub hr: HomophoneReplacerConfig,
}
```

--------------------------------

### OfflinePunctuation::create

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflinePunctuation.html

Creates a new instance of the OfflinePunctuation struct using the provided configuration.

```APIDOC
## pub fn create(config: &OfflinePunctuationConfig) -> Option<Self>

### Description
Creates an offline punctuator instance from the provided `OfflinePunctuationConfig`.

### Parameters
- **config** (&OfflinePunctuationConfig) - Required - The configuration object used to initialize the punctuator.

### Returns
- **Option<Self>** - Returns an instance of `OfflinePunctuation` if initialization is successful, otherwise `None`.
```

--------------------------------

### Define OnlineParaformerModelConfig struct

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlineParaformerModelConfig.html

The structure definition for configuring Online Paraformer models with optional encoder and decoder paths.

```rust
pub struct OnlineParaformerModelConfig {
    pub encoder: Option<String>,
    pub decoder: Option<String>,
}
```

--------------------------------

### Define OfflinePunctuationConfig struct

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflinePunctuationConfig.html

The structure definition for configuring offline punctuation models.

```rust
pub struct OfflinePunctuationConfig {
    pub model: OfflinePunctuationModelConfig,
}
```

--------------------------------

### Define OfflineSpeakerSegmentationPyannoteModelConfig

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineSpeakerSegmentationPyannoteModelConfig.html

The struct definition for configuring the Pyannote segmentation model path.

```rust
pub struct OfflineSpeakerSegmentationPyannoteModelConfig {
    pub model: Option<String>,
}
```

--------------------------------

### OfflineMoonshineModelConfig Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineMoonshineModelConfig.html

The structure definition for configuring Moonshine models, containing optional paths for preprocessor, encoder, and various decoder components.

```rust
pub struct OfflineMoonshineModelConfig {
    pub preprocessor: Option<String>,
    pub encoder: Option<String>,
    pub uncached_decoder: Option<String>,
    pub cached_decoder: Option<String>,
    pub merged_decoder: Option<String>,
}
```

--------------------------------

### OfflineSpeechDenoiser::sample_rate

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineSpeechDenoiser.html

Retrieves the expected sample rate for the denoiser.

```APIDOC
## pub fn sample_rate(&self) -> i32

### Description
Returns the model sample rate expected by this denoiser instance.

### Returns
- **i32** - The expected sample rate in Hz.
```

--------------------------------

### FastClusteringConfig Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.FastClusteringConfig.html

The structure definition for FastClusteringConfig containing the configuration fields for clustering.

```rust
pub struct FastClusteringConfig {
    pub num_clusters: i32,
    pub threshold: f32,
}
```

--------------------------------

### DisplayManager::finalize_sentence

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.DisplayManager.html

Moves the current partial text into the list of finalized sentences.

```APIDOC
## pub fn finalize_sentence(&mut self)

### Description
Moves the current partial text into the finalized sentence list with a timestamp.
```

--------------------------------

### OfflineStream Methods

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineStream.html

Methods available for interacting with an OfflineStream instance.

```APIDOC
## pub fn accept_waveform(&self, sample_rate: i32, samples: &[f32])

### Description
Appends audio samples to the stream for processing.

### Parameters
- **sample_rate** (i32) - The sample rate of the audio data.
- **samples** (&[f32]) - A slice containing the audio samples.

## pub fn get_result(&self) -> Option<OfflineRecognizerResult>

### Description
Fetches the current recognition result from the stream.

### Returns
- **Option<OfflineRecognizerResult>** - The recognition result if available.

## pub fn set_option(&self, key: &str, value: &str)

### Description
Sets an option for the stream.

## pub fn get_option(&self, key: &str) -> String

### Description
Retrieves the value of a specific option.

## pub fn has_option(&self, key: &str) -> bool

### Description
Checks if a specific option exists for the stream.
```

--------------------------------

### DenoisedAudio Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.DenoisedAudio.html

The structure used to store denoised audio output from offline or online denoisers.

```rust
pub struct DenoisedAudio {
    pub samples: Vec<f32>,
    pub sample_rate: i32,
}
```

--------------------------------

### AudioTaggingOfflineStream Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.AudioTaggingOfflineStream.html

The structure definition for the offline audio tagging input stream.

```rust
pub struct AudioTaggingOfflineStream { /* private fields */ }
```

--------------------------------

### DisplayManager Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.DisplayManager.html

The internal structure definition for DisplayManager.

```rust
pub struct DisplayManager { /* private fields */ }
```

--------------------------------

### Define OnlineTransducerModelConfig struct

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlineTransducerModelConfig.html

The structure used to configure online transducer models with optional paths for the encoder, decoder, and joiner.

```rust
pub struct OnlineTransducerModelConfig {
    pub encoder: Option<String>,
    pub decoder: Option<String>,
    pub joiner: Option<String>,
}
```

--------------------------------

### TenVadModelConfig Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.TenVadModelConfig.html

The structure definition for TenVadModelConfig, outlining the available configuration fields for VAD models.

```rust
pub struct TenVadModelConfig {
    pub model: Option<String>,
    pub threshold: f32,
    pub min_silence_duration: f32,
    pub min_speech_duration: f32,
    pub window_size: i32,
    pub max_speech_duration: f32,
}
```

--------------------------------

### OnlinePunctuationConfig Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlinePunctuationConfig.html

The primary structure used to configure online punctuation, containing an instance of OnlinePunctuationModelConfig.

```rust
pub struct OnlinePunctuationConfig {
    pub model: OnlinePunctuationModelConfig,
}
```

--------------------------------

### VadModelConfig Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.VadModelConfig.html

The structure definition for VadModelConfig, containing fields for Silero and Ten VAD models, sample rate, thread count, provider, and debug mode.

```rust
pub struct VadModelConfig {
    pub silero_vad: SileroVadModelConfig,
    pub ten_vad: TenVadModelConfig,
    pub sample_rate: i32,
    pub num_threads: i32,
    pub provider: Option<String>,
    pub debug: bool,
}
```

--------------------------------

### SpokenLanguageIdentificationWhisperConfig Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.SpokenLanguageIdentificationWhisperConfig.html

The definition of the configuration struct for Whisper-based spoken language identification.

```rust
pub struct SpokenLanguageIdentificationWhisperConfig {
    pub encoder: Option<String>,
    pub decoder: Option<String>,
    pub tail_paddings: i32,
}
```

--------------------------------

### OfflineRecognizer::create

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineRecognizer.html

Creates a new instance of an OfflineRecognizer based on the provided configuration.

```APIDOC
## pub fn create(config: &OfflineRecognizerConfig) -> Option<Self>

### Description
Creates a recognizer from the provided configuration.

### Parameters
- **config** (&OfflineRecognizerConfig) - Required - The configuration object for the recognizer.

### Returns
- **Option<Self>** - Returns an instance of OfflineRecognizer if successful, otherwise None.
```

--------------------------------

### SpokenLanguageIdentificationConfig struct definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.SpokenLanguageIdentificationConfig.html

The primary configuration structure used for initializing spoken language identification.

```rust
pub struct SpokenLanguageIdentificationConfig {
    pub whisper: SpokenLanguageIdentificationWhisperConfig,
    pub num_threads: i32,
    pub debug: bool,
    pub provider: Option<String>,
}
```

--------------------------------

### VoiceActivityDetector::front

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.VoiceActivityDetector.html

Retrieves the oldest speech segment from the queue.

```APIDOC
## pub fn front(&self) -> Option<SpeechSegment>

### Description
Returns the front speech segment from the internal queue, if one is available.
```

--------------------------------

### LinearResampler::input_sample_rate

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.LinearResampler.html

Retrieves the input sample rate.

```APIDOC
## pub fn input_sample_rate(&self) -> i32

### Description
Returns the input sample rate configured for this resampler.

### Returns
- **i32** - The input sample rate in Hz.
```

--------------------------------

### Define OfflineSpeechDenoiserDpdfNetModelConfig struct

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineSpeechDenoiserDpdfNetModelConfig.html

The configuration struct containing the optional model path for DPDFNet.

```rust
pub struct OfflineSpeechDenoiserDpdfNetModelConfig {
    pub model: Option<String>,
}
```

--------------------------------

### VoiceActivityDetector::flush

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.VoiceActivityDetector.html

Flushes buffered speech into the output queue.

```APIDOC
## pub fn flush(&self)

### Description
Forces the detector to flush any remaining buffered trailing speech into the output queue.
```

--------------------------------

### LinearResampler::output_sample_rate

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.LinearResampler.html

Retrieves the output sample rate.

```APIDOC
## pub fn output_sample_rate(&self) -> i32

### Description
Returns the output sample rate configured for this resampler.

### Returns
- **i32** - The output sample rate in Hz.
```

--------------------------------

### OfflineTtsKokoroModelConfig Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineTtsKokoroModelConfig.html

The structure definition for configuring Kokoro TTS models, including optional paths for model components and scaling factors.

```rust
pub struct OfflineTtsKokoroModelConfig {
    pub model: Option<String>,
    pub voices: Option<String>,
    pub tokens: Option<String>,
    pub data_dir: Option<String>,
    pub length_scale: f32,
    pub dict_dir: Option<String>,
    pub lexicon: Option<String>,
    pub lang: Option<String>,
}
```

--------------------------------

### OfflineTtsMatchaModelConfig Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineTtsMatchaModelConfig.html

The structure definition for configuring the Matcha TTS model.

```rust
pub struct OfflineTtsMatchaModelConfig {
    pub acoustic_model: Option<String>,
    pub vocoder: Option<String>,
    pub lexicon: Option<String>,
    pub tokens: Option<String>,
    pub data_dir: Option<String>,
    pub noise_scale: f32,
    pub length_scale: f32,
    pub dict_dir: Option<String>,
}
```

--------------------------------

### OnlineSpeechDenoiser::frame_shift_in_samples

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlineSpeechDenoiser.html

Retrieves the preferred input frame shift.

```APIDOC
## pub fn frame_shift_in_samples(&self) -> i32

### Description
Returns the preferred input frame shift, measured in samples.

### Returns
- **i32** - The frame shift in samples.
```

--------------------------------

### Define OfflineTransducerModelConfig struct

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineTransducerModelConfig.html

The structure definition for configuring transducer-style models.

```rust
pub struct OfflineTransducerModelConfig {
    pub encoder: Option<String>,
    pub decoder: Option<String>,
    pub joiner: Option<String>,
}
```

--------------------------------

### OfflineRecognizer::create_stream

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineRecognizer.html

Creates an empty offline stream for processing audio data.

```APIDOC
## pub fn create_stream(&self) -> OfflineStream

### Description
Creates an empty offline stream.

### Returns
- **OfflineStream** - A new stream instance.
```

--------------------------------

### OfflineSpeechDenoiserModelConfig Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineSpeechDenoiserModelConfig.html

The structure used to aggregate model configurations for the OfflineSpeechDenoiser. Configure exactly one model family when initializing.

```rust
pub struct OfflineSpeechDenoiserModelConfig {
    pub gtcrn: OfflineSpeechDenoiserGtcrnModelConfig,
    pub dpdfnet: OfflineSpeechDenoiserDpdfNetModelConfig,
    pub num_threads: i32,
    pub debug: bool,
    pub provider: Option<String>,
}
```

--------------------------------

### OnlineSpeechDenoiserConfig Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OnlineSpeechDenoiserConfig.html

The primary structure used to configure the online speech denoiser.

```rust
pub struct OnlineSpeechDenoiserConfig {
    pub model: OfflineSpeechDenoiserModelConfig,
}
```

--------------------------------

### OfflineModelConfig Struct Definition

Source: https://docs.rs/sherpa-onnx/latest/sherpa_onnx/struct.OfflineModelConfig.html

The structure definition containing fields for various supported offline ASR models and global configuration parameters.

```rust
pub struct OfflineModelConfig {Show 26 fields
    pub transducer: OfflineTransducerModelConfig,
    pub paraformer: OfflineParaformerModelConfig,
    pub nemo_ctc: OfflineNemoEncDecCtcModelConfig,
    pub whisper: OfflineWhisperModelConfig,
    pub tdnn: OfflineTdnnModelConfig,
    pub sense_voice: OfflineSenseVoiceModelConfig,
    pub moonshine: OfflineMoonshineModelConfig,
    pub fire_red_asr: OfflineFireRedAsrModelConfig,
    pub dolphin: OfflineDolphinModelConfig,
    pub zipformer_ctc: OfflineZipformerCtcModelConfig,
    pub canary: OfflineCanaryModelConfig,
    pub wenet_ctc: OfflineWenetCtcModelConfig,
    pub omnilingual: OfflineOmnilingualAsrCtcModelConfig,
    pub medasr: OfflineMedAsrCtcModelConfig,
    pub funasr_nano: OfflineFunASRNanoModelConfig,
    pub fire_red_asr_ctc: OfflineFireRedAsrCtcModelConfig,
    pub qwen3_asr: OfflineQwen3ASRModelConfig,
    pub cohere_transcribe: OfflineCohereTranscribeModelConfig,
    pub tokens: Option<String>,
    pub num_threads: i32,
    pub debug: bool,
    pub provider: Option<String>,
    pub model_type: Option<String>,
    pub modeling_unit: Option<String>,
    pub bpe_vocab: Option<String>,
    pub telespeech_ctc: Option<String>,
}
```
Tokens:
Raw