Our friends over at Simli put a face to our Aura voices to bring our AI Agent to life using their Audio-to-Video API. Looking good! 😎
Faceless voice agents became a bit boring. So, I've put together this demo integrating Simli's Realtime Audio-to-Video API with Deepgram's AI Agent Tech Demo! Check it out on GitHub: https://lnkd.in/emv6CcnZ
Yu He
2mo
Hey, are you familiar with tweaking the pace of your text-to-speech API to get the perfect flow?
Solving the bike routing problem for fun and profit
3moHmm. This is pretty good -certainly sophisticated. I like how the face moves while waiting for the voice response. For what use-cases would this tech at the current stage of development provide a plus? Curious about your thoughts. To be frank the TTS cadence and intonation isn't great, and adding a face in the uncanny valley doesn't help. If I can't use lip movement to aide in my understanding of speech, how is it helpful?