For when you just want to read or write data from or to some audio hardware.
use render_callback::{AudioBuffers, Backend, CurrentPlatformBackend};
fn main() {
let backend = CurrentPlatformBackend::new().unwrap();
let mut session = {
let input_device = backend.default_input_device().unwrap();
let output_device = backend.default_output_device().unwrap();
// Or, use backend.all_devices() to iterate over available devices and pick the ones you want
backend.start_session(
44100.0, // Sample rate
input_device,
output_device,
Box::new(|input, output| {
// Just a &[f32] with one sample per input channel interleaved
let interleaved_inputs = input.interleaved_frames();
let num_input_channels = input.num_channels();
let num_input_frames = input.num_frames(); // interleaved_inputs.len() / num_input_channels
// Just a &mut [f32] with one sample per output channel interleaved
let interleaved_outputs = output.interleaved_frames_mut();
let num_output_channels = output.num_channels();
let num_output_frames = output.num_frames(); // interleaved_outputs.len() / num_output_channels
// The API guarantees that there are an equal amount of input frames
// and output frames, but the channel counts might differ.
assert_eq!(num_input_frames, num_output_frames);
// Do stuff with them
})
).unwrap()
};
// Do other stuff in your application. Maybe set up an MPSC channel to
// communicate with the real-time thread?
// Audio will stop when you drop the session object, so make sure to keep it
// alive as long as you need.
}Look at src/traits.rs for the complete API.