Listening Ahead Ever so Slightly

Chris Chafe1, Mike Dickey2
1Stanford University, Stanford, CA 94305 USA
2JackTrip Labs, Palo Alto, CA 94304 USA

Regulator is a PLC algorithm implemented in JackTrip 3.0 and the experimental webtrip webapp for browser / mobile clients. Its role is reducing latency for interactive network audio applications while maintaining uncompressed audio streaming quality. The paper has been accepted for upcoming publication in JONMA (2026).

Listening Ahead Ever so Slightlyr presents a novel packet loss concealment system called “Regulator” designed for interactive network audio applications, particularly server-mediated “jam rooms” in the cloud where large ensembles of musicians perform together synchronously. The system addresses the critical challenge of maintaining continuous high-quality audio output while minimizing latency penalties in unreliable network environments including WiFi. The Regulator architecture combines four core components: a central Regulator class orchestrating packet loss concealment operations, a BurgAlgorithm class implementing maximum entropy autoregressive prediction for missing audio samples, a Channel class providing per-channel independent modeling for multi-channel streams, and a RegulatorWorker class offering asynchronous processing through lock-free threading. The system’s packet selection logic evaluates multiple candidate packets for optimal playback timing and its cross-fading mechanisms ensure smooth transitions between predicted and real audio to minimize audible artifacts. An adaptive tolerance management system automatically adjusts buffer headroom based on observed network jitter patterns. Unlike traditional approaches that add significant delay or produce glitchy substitutes, Regulator passes received streams through with minimal added delay while providing high-quality packet loss concealment supporting various audio formats including different sample rates, bit depths, frame sizes, and multi-channel configurations for deployment across heterogeneous network audio scenarios.