Struct AudioResampler 

pub struct AudioResampler { /* private fields */ }

Implementations

impl AudioResampler

pub fn new( in_rate: u32, out_rate: u32, channels: u8, chunk_size: usize, ) -> Result<Self, AudioError>

Construct a resampler for in_rate → out_rate with channels channels processing chunk_size input frames per call. Returns an error if any rate is zero.

pub fn in_rate(&self) -> u32

pub fn out_rate(&self) -> u32

pub fn channels(&self) -> u8

pub fn chunk_size(&self) -> usize

pub fn process( &mut self, frame: &AudioFrame, out: &mut Vec<f32>, ) -> Result<(), AudioError>

Process frame.samples (interleaved) and append output samples (interleaved) into out. Carries any partial input chunk internally for the next call.

The output PTS is the same as the input PTS — the resampler itself doesn’t expose its internal lookahead in our wire model (the encoder converts that into pre_skip ticks at the file header level).

pub fn flush(&mut self, out: &mut Vec<f32>) -> Result<(), AudioError>

Flush any carry by zero-padding to a full chunk and processing it. Useful at end-of-stream to drain the rubato sinc filter.