VibeCode
Log in
cpaua·1 min186

Voxtral: Open-Weights TTS Alternative to ElevenLabs by Mistral

Читати українською

An open alternative to ElevenLabs with open weights has appeared.

Voxtral is a speech synthesis (text-to-speech) model from Mistral:

- only 4 billion parameters
- 70 ms latency for voice agents
- voice cloning from 3 seconds of audio
- 9 languages + cross-lingual transfer
- 68.4% wins compared to ElevenLabs Flash v2.5

Open weights are Hugging Facemistralai/Voxtral-4B-TTS-2603huggingface.co/mistralai/Voxtral-4B-TTS-2603 on Hugging Face.

Share:
Author
cpaua

VibeCode blog admin. Writing about vibe coding, AI and open source.

Comments

To leave a comment, log in or sign up
Loading...

Related articles

How to Get Free Claude Opus 4.7 and GPT 5.5 Codex via FreeModel

Step-by-step guide to register on freemodel.dev, verify via Telegram, create API keys, and use Claude Code or Codex with a free weekly $66 limit.

Claude Opus 4.8 Released: Better Reasoning, Cheaper Fast Mode

Anthropic launches Claude Opus 4.8 with stronger reasoning, efficient tool calling, Dynamic Workflows in Claude Code, Effort Control, and 3x cheaper Fast mode.

Claude Code Updates: Faster Streaming, Reliable MCP, Self-Healing Sessions

Latest Claude Code improvements: fullscreen renderer fixes, real-time streaming, clearer errors, faster compaction, more reliable MCP, self-healing sessions, easier feedback.