Local and Open Source Speech to Speech Assistant

In this video, I'll walk you through how to set up a completely local voice assistant using my project, Verbi. We'll configure three local API endpoints: Fast Whisper for speech to text, OLAMMA for the language model, and Mello TTS for text to speech. Make sure to check out my previous videos for initial setup instructions and enjoy experimenting with state-of-the-art speech components!

LINKS:
Verbi Videos:

💻 RAG Beyond Basics Course:

Let's Connect:

Signup for Newsletter, localgpt:

TIMESTAMP:
00:00 Introduction to Verbi
01:15 Setting Up Local Models
02:56 Configuring Fast Whisper API
04:41 Installing Mello TTS
08:47 Running Verbi and Testing
12:36 Conclusion and Future Updates

All Interesting Videos:

[mathiasgentech]

I'm looking at the github repo and it's super easy to follow. The modular nature is nice. I had cycled between using speechrecognizer with vosk, then speechrecognizer with a local whisper, then realtime transcription with whisper. But I kept having to duct tape solutions for problems that are easily solved by separating it out. Verbi helped fill in some gaps I hadn't considered. I'll be sure to leave a note in my code's readme referencing your project.

[RickySupriyadi]

wow that Local TTS sound so natural, really cool

[kryptonic]

Great example. How can you integrate a RAG engine to accept word and excel files?

[bizmark]

Is the project "verbi" still active?

[preben]

How hard would it be to add rag support to this?

[brto]

The TTS sound a bit robotic but I like ❤ Appreciate the hard work. Keep it 100 💪

[SAINGSAB]

Have anyone of us here building the TTS on our own local languages beside the English?