Hello everyone,
I was checking out the pricing page and noticed ASR (Automatic Speech Recognition) and TTS (Text-to-Speech) listed, but I couldn’t find any information on the real-time service that '@phoney ai' is using through Groq. Has anyone here come across this service or knows more about what Groq offers in terms of real-time capabilities? Any help or insights would be greatly appreciated!
Hi Aliep!
We have Whisper transcription models, and a TTS model as separate model endpoints. Phonely builds on top of Whisper and adds their own coordination/infrastructure on top of it. Unfortunately some of that is their secret sauce, and we don’t know how they’ve implemented it.
We have an integrations page on how to use Groq + LiveKit, an open source real-time voice platform: https://console.groq.com/docs/livekit
Over the coming weeks we’ll be releasing more demos with LiveKit, VAPI, and PipeCat, as well as completely vanilla implementations as well.