realtime-audio

Tweaks and config for modep-host / modep-ui for Raspberry Pi 4/400.

Live multi-effect guitar pedalboard in C++17 for the pi-Stomp (Raspberry Pi 5): 30+ JUCE-DSP effects, Neural Amp Modeler, lock-free RCU chain editing, LVGL on-device UI + Svelte/SSE web control

Phone-callable voice agent: Vobiz XML <Stream> bridged to the Gemini Live API. No audio resampled anywhere, real barge-in, and a mock client that holds a whole conversation without placing a call.

GenPark AI Agent Skill - Phonetic Soundex and acoustic confusion matrix resolver correcting domain-specific STT transcription errors.

Jacob Collier-like harmonizer, because I'm jealous and I want a choir for myself too

WebSocket Speechmatics Flow bridge for AVR: receives PCM audio from avr-core and streams STS audio responses in real time.

Audio PeakMeter for JackAudio

GenPark AI Agent Skill - Real-time conversational voice activity detection, dynamic silence endpointing and barge-in interruption arbitrator.

High-performance music engine for .NET with SIMD-accelerated operations. Fast note manipulation, chord analysis, and musical theory computations

Low-latency streaming prosody, SSML, and emotion pacing modulator dynamically tailoring speech pitch, cadence, and pause inflections.

macOS app that measures and trains musical timing from MIDI. Swift, 926 tests, one self-checking quality gate. Built with Claude Code under my direction and verification.

GenPark AI Agent Skill - Real-time RTP and WebSocket audio packet jitter buffer optimizer smoothing network latency and clock drift for speech streaming.

GenPark AI Agent Skill - PSTN/SIP voice telephony state machine managing DTMF tone decoders, call transfer handoffs and IVR navigation.

Enterprise Realtime Voice AI Technical Interviewer — Adaptive System Design & Resume Deep-Dive with Deepgram Nova-2, OpenAI & ElevenLabs.

Real-time pitch correction with graphical note editing, Auto-Key and four-voice harmony. VST3 + standalone for Windows. Built with JUCE.

End-to-end conversational voice agent latency telemetry profiler tracking VAD, STT, LLM-TTFT, TTS-TTFB, and playout bottlenecks.

Realtime RVC voice changer for AMD Ryzen AI (XDNA 2 NPU) - ONNX Runtime, VitisAI EP, DirectML, CPU; C++20 + ImGui, Python model conversion tools

A high-performance, GPU-optimized real-time speech-to-text (STT) streaming server built with WebSocket support for multiple concurrent clients. This project leverages the Kyutai STT model and is optimized for NVIDIA RTX 4090 GPUs, providing low-latency transcription for audio streams.

Real-time microphone processing for macOS — EQ, compressor, noise gate, voice isolation, reverb, delay, pitch, LUFS analysis, take recording. For podcasters, streamers and gamers.

Low-latency osu! hitsounds for Linux. Per-key samples fired the instant you press, for osu!, mania, and other rhythm games.