Struct Model 

pub struct Model { /* private fields */ }

Streaming zipformer transducer speech recognition model.

Implementations

impl Model

pub async fn from_pretrained<S: AsRef<str>>(model: S) -> Result<Config>

Available on crate feature download-models only.

Create a Config from a pretrained model on huggingface.

use sherpa_transducers::asr;

let model = asr::Model::from_pretrained("nytopop/nemo-conformer-transducer-en-80ms")
    .await?
    .build()?;

pub async fn from_pretrained_arch<S>( model: S, arch: Arch, tokens: S, ) -> Result<Config>

where S: AsRef<str>,

Available on crate feature download-models only.

Create a Config from a pretrained ASR model on huggingface without a config.json.

pub fn online_stream(&self) -> Result<OnlineStream>

Make an OnlineStream for incremental speech recognition.

pub fn phased_stream(&self, n_phase: usize) -> Result<PhasedStream>

Make a PhasedStream for incremental speech recognition.

Trades off increased compute utilization for lower latency transcriptions (sub chunk size).

pub fn sample_rate(&self) -> usize

Returns the native sample rate.

pub fn chunk_size(&self) -> usize

Returns the chunk size at the native sample rate.

Trait Implementations

impl Clone for Model

fn clone(&self) -> Model

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.