OpenAI Voice-to-Voice Integration
Budget: ₹600 – ₹1,500 INR
I have a very small Python script (one page) from the Pipecat examples that uses Telnyx to place an outbound call: https://github.com/pipecat-ai/pipecat-examples/tree/main/telnyx-chatbot/outbound.
Right now the flow is speech-to-text → AI processing → text-to-speech. I want that entire chain replaced with OpenAI’s new voice-to-voice model so the call can stream audio back and forth directly, no interim transcription or synthesis steps required.
All existing AI processing features can go; the goal is a clean swap to voice-to-voice while keeping the outbound-call logic that’s already wired to Telnyx. The codebase is tiny, so this should mostly involve updating the command only because rest pipecat will handle
Right now the flow is speech-to-text → AI processing → text-to-speech. I want that entire chain replaced with OpenAI’s new voice-to-voice model so the call can stream audio back and forth directly, no interim transcription or synthesis steps required.
All existing AI processing features can go; the goal is a clean swap to voice-to-voice while keeping the outbound-call logic that’s already wired to Telnyx. The codebase is tiny, so this should mostly involve updating the command only because rest pipecat will handle