#pipecat — Public Fediverse posts

- Requires Gemini API & Daily API keys
- Supports Web, React, iOS, Android & more
- Server runs #Pipecat pipeline for orchestration

💡 Architecture Benefits:
- Smart proxy approach for optimal performance
- Server-side Python logic customization

Building Voice #AI Applications: #Gemini Multimodal Live API with #WebRTC and #Pipecat 🎯

🧵

🔧 Simple web app using #opensource #PipecatSDK integrates with Gemini API through WebRTC for voice communication

🌐 #WebRTC advantages over WebSockets:

📝 If you missed my talk at #ODSCWest yesterday, don't worry! The writeup is ready ✨

1️⃣ Let's see the challenges of building voice bots that actually sound human 💁‍♀️ https://www.zansara.dev/posts/2024-09-05-building-voice-agents-with-open-source-tools-part-1/

2️⃣ Let's make a simple #Pipecat voice bot with the same latency as #GPT4o #VoiceMode for a fraction of the cost 💸 https://www.zansara.dev/posts/2024-10-30-building-voice-agents-with-open-source-tools-part-2/

#AI #GenAI #LLM #GPT #VoiceAgent #Chatbot #ODSCWest

Thank you to everyone that just attended my virtual tutorial session at #ODSC Europe!

We talked about the state of the art of voice agents in the age of #GenAI and #LLMs, we implemented a little bot with #Pipecat and then tested out an innovative approach to use system prompts to control the conversation without using huge and overly complicated system prompts.

You can find all the material here: https://www.zansara.dev/talks/2024-09-05-odsc-europe-voice-agents/

Stay tuned for more content 📚