Struct AudioConfig

pub struct AudioConfig {
    pub sampling_rate: usize,
    pub frame_rate: f64,
    pub audio_encoding_config: AudioSpectrogramConfig,
    pub chunk_length_s: Option<f64>,
}

Configuration for audio processing and tokenization.

This struct contains all parameters needed to process audio files and convert them into token sequences that can be mixed with text tokens.

Fields

• sampling_rate - Target sampling rate in Hz (e.g., 16000)
• frame_rate - Number of frames per second for the tokenizer model
• audio_encoding_config - Spectrogram generation parameters
• chunk_length_s - Optional chunk length in seconds for padding

Fields

sampling_rate: usize

frame_rate: f64

audio_encoding_config: AudioSpectrogramConfig

chunk_length_s: Option<f64>

Implementations

impl AudioConfig

pub fn new( sampling_rate: usize, frame_rate: f64, encoding_config: AudioSpectrogramConfig, chunk_length_s: Option<f64>, ) -> Result<Self>

Creates a new AudioConfig with validation.

Arguments

• sampling_rate - Target sampling rate in Hz (must be > 0)
• frame_rate - Number of frames per second (must be > 0)
• encoding_config - Spectrogram configuration
• chunk_length_s - Optional chunk length in seconds (must be > 0 if provided)

Returns

A new AudioConfig instance.

Errors

Returns an error if any parameter is invalid.

pub fn chunk_frames(&self) -> Result<usize>

Calculates the number of audio frames per chunk.

Returns

The number of frames per chunk based on chunk length and sampling rate.

Errors

Returns an error if chunk_length_s is not set.

pub fn audio_length_per_tok(&self) -> usize

Calculates the length of audio (in samples) represented by each token.

This determines the downsampling factor from audio samples to tokens based on the frame rate and spectrogram hop length.

Returns

Number of audio samples per token.

Trait Implementations

impl Clone for AudioConfig

fn clone(&self) -> AudioConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for AudioConfig

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for AudioConfig

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for AudioConfig

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.