Interactive voice agents—from automated clinic receptionists and customer support phone lines to live interview copilots and face-to-face translators—require sub-300ms turn-taking latency, barge-in interruption handling, and deterministic tool calling.
If you are looking to hire a freelance OpenAI Realtime API, Whisper, and Voice AI agent programmer, I am Aldian Fazrihady—creator of Interview Copilot, Transcribe Copilot, and Face-to-Face Translator.
Voice AI & Audio Engineering Capabilities
- WebRTC & WebSocket Audio Streaming: Browser `AudioWorklet` PCM streaming, server-side Voice Activity Detection (VAD), and instant barge-in cancellation.
- Embedded Hardware & Edge Voice Control: Huddle Voice Control Integration: Processing Audio Telemetry on Remote Embedded Hardware
- Cost-Optimized Batch & Streaming Pipelines: Reducing OpenAI Bills by 50% with the OpenAI Batch API