OpenAudio S1 Mini TTS LOCAL Test & Install (Multilingual Voice Cloning!)

Timestamps:

00:00 - Intro
00:48 - First Look
03:41 - Local Install Guide
06:27 - Initial Testing
10:26 - Long Context Testing
12:45 - Voice Cloning Test
15:45 - Multilingual Cloning Test
17:47 - Random Testing
19:17 - Speech Control Test
20:26 - Closing Thoughts

AI Integration Consulting: https://bijanbowen.com
Join the AI Discord:   / discord  

In this video, we take an in-depth look at S1 Mini, a new open-source TTS model released by OpenAudio (formerly Fish Audio). Built on a Qwen3-based architecture, this model is designed for multilingual voice cloning, speech control, and long-form generation — all runnable locally.

We start by exploring the model's structure and capabilities, then walk through the full local installation process. Once set up, we perform a range of tests including basic synthesis, long context generation, and voice cloning across multiple languages. The model also supports speech control tags, which allow for expressive and stylistic control over output.

GitHub Repo: https://github.com/fishaudio/fish-speech