Twilio ElevenLabs Assistant Integration
Budget: ₹1,500 – ₹12,500 INR
I want to wire Twilio’s telephony, ElevenLabs’ text-to-speech, and OpenAI’s chat completion into a single, seamless voice interface that callers can use as a professional-sounding personal assistant. When someone rings the Twilio number, their speech should be transcribed, sent to the GPT model for reasoning, and the reply should come back through ElevenLabs so the caller hears a natural voice.
For this first milestone the only skill the assistant needs is answering general questions accurately and in a polished, business-like tone—no scheduling or weather features yet. I need clean, well-commented code (Python or Node.js—your choice as long as the libraries are well-supported) that:
• receives inbound Twilio voice calls,
• streams the audio, handles the STT step,
• passes the text prompt to OpenAI,
• feeds the model’s response to ElevenLabs for voice synthesis, and
• delivers the audio back to the caller in real time without noticeable lag.
A small README with environment variable setup, API key handling, and a quick “run locally” guide will be enough for documentation, and I’ll test everything on my own Twilio and ElevenLabs accounts. If you have suggestions for latency reduction, caching, or prompt engineering tricks to keep answers succinct and professional, please include them in the hand-off.
For this first milestone the only skill the assistant needs is answering general questions accurately and in a polished, business-like tone—no scheduling or weather features yet. I need clean, well-commented code (Python or Node.js—your choice as long as the libraries are well-supported) that:
• receives inbound Twilio voice calls,
• streams the audio, handles the STT step,
• passes the text prompt to OpenAI,
• feeds the model’s response to ElevenLabs for voice synthesis, and
• delivers the audio back to the caller in real time without noticeable lag.
A small README with environment variable setup, API key handling, and a quick “run locally” guide will be enough for documentation, and I’ll test everything on my own Twilio and ElevenLabs accounts. If you have suggestions for latency reduction, caching, or prompt engineering tricks to keep answers succinct and professional, please include them in the hand-off.