Struct InstrumentedTranscriptionModel 

pub struct InstrumentedTranscriptionModel {
    pub inner: Arc<dyn TranscriptionModel>,
    pub alias: String,
    pub provider_id: String,
    pub timeout: Option<Duration>,
    pub retry: Option<RetryConfig>,
}

Wrapper around a TranscriptionModel adding timeout / retry / metrics to the batch transcribe call.

The timeout applies batch-wide — operators tuning timeout for batched ingest need to budget for the full batch, not a per-item slice.

Fields

inner: Arc<dyn TranscriptionModel>

alias: String

provider_id: String

timeout: Option<Duration>

retry: Option<RetryConfig>

Trait Implementations

impl ModelInfo for InstrumentedTranscriptionModel

fn model_id(&self) -> &str

The underlying model identifier (e.g. a HuggingFace repo ID or API model name).

fn active_execution_providers(&self) -> Vec<String>

Names of the ONNX Runtime execution providers this session was built with, in priority order. Empty for remote and non-ONNX models.

impl TranscriptionModel for InstrumentedTranscriptionModel

fn transcribe<'life0, 'async_trait>( &'life0 self, audios: Vec<AudioInput>, options: TranscribeOptions, ) -> Pin<Box<dyn Future<Output = Result<Vec<TranscribeResult>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Transcribe a batch of audio inputs.

fn supported_languages(&self) -> &[String]

ISO 639-1 language codes the model can handle. May be empty for providers that report language only at runtime via auto-detection.

fn warmup<'life0, 'async_trait>( &'life0 self, ) -> Pin<Box<dyn Future<Output = Result<()>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Optional warmup hook. The default is a no-op.

fn transcribe_one<'life0, 'async_trait>( &'life0 self, audio: AudioInput, options: TranscribeOptions, ) -> Pin<Box<dyn Future<Output = Result<TranscribeResult>> + Send + 'async_trait>>

where Self: Sync + 'async_trait, 'life0: 'async_trait,

Transcribe a single audio stream — convenience over transcribe.