rift-media

Part of the riftd project — serverless P2P voice + text chat over UDP.

What's in this crate?

rift-media handles everything audio:

• Audio Capture — Cross-platform microphone input via cpal
• Audio Playback — Cross-platform speaker output via cpal
• Opus Codec — High-quality, low-latency voice encoding
• VAD — Voice activity detection to reduce bandwidth
• Level Metering — Audio level detection for UI indicators
• Quality Presets — Low/medium/high bitrate configurations

Usage

use rift_media::{AudioConfig, OpusEncoder, OpusDecoder, AudioCapture};

let config = AudioConfig::default();

// Encoding
let mut encoder = OpusEncoder::new(&config)?;
let encoded = encoder.encode_f32(&samples, &mut output)?;

// Decoding
let mut decoder = OpusDecoder::new(&config)?;
let decoded = decoder.decode_f32(&encoded, &mut output)?;

// Capture from microphone
let capture = AudioCapture::new(&config)?;
capture.start(|samples| { /* process samples */ })?;

Related Crates

Citation

If you use riftd in academic work, please cite:

de Beer, R. (2026). Predictive Rendezvous: Time–Intent–Deterministic Peer Coordination Without Infrastructure. Zenodo. https://doi.org/10.5281/zenodo.18528430

License

Licensed under either of Apache License, Version 2.0 or MIT license at your option.