Subscribe
Sign in
Fish Speech 1.4 — Open-Source Multilingual Text-to-Speech with Voice Cloning
Your Voice, Your Way: Open-Source TTS
Powerful, fast, and natural speech in any language. Clone voices instantly. Self-host or use our service. Lightning-fast, affordable pricing.
17
Replies
Yue Leng
Maker
📌
Excited to introduce Fish Audio 1.4 - now open-source and more powerful than ever! 🎉
What's new:
- Trained on 700k hours of multilingual data (up from 200k)
- Now supports 8 languages: English, Chinese, German, Japanese, French, Spanish, Korean, and Arabic
- Fully open-source, empowering developers and researchers worldwide
Our mission: Make cutting-edge voice tech accessible to everyone.
Key features:
- Lightning-fast TTS with ultra-low latency
- Instant voice cloning
- Self-host or use our cloud service
- Simple, flat-rate pricing
Try it out:
- Playground: https://fish.audio
- GitHub: https://github.com/fishaudio/fis...
- HuggingFace Model: https://huggingface.co/fishaudio...
- Demo: https://huggingface.co/spaces/fi...
We can't wait to see what you'll create with Fish Audio. Happy voice building! 🎧🐠
Share
@lengyue This is amazing, are you planning to add Hindi to the languages list? there's a huge market with hindi audio, pls do explore
@lengyue Congratulations on the launch of Fish Audio 1.4! 🎉 It's incredible to see the platform grow with 700k hours of multilingual data and support for 8 languages—this is a huge step forward! Making it open-source will truly empower developers and researchers across the globe. Excited to see the innovations that come from this. Keep up the amazing work!
Congratulations on launching such an innovative product! I'm really intrigued by the idea of having powerful, fast, and natural speech synthesis available in any language—it's a game changer for accessibility and creativity.
The feature that stands out to me is the ability to clone voices instantly. This opens up so many possibilities for content creators and developers alike. Additionally, the option to self-host or use your service provides flexibility that many users will appreciate.
I’m curious about how you handle voice cloning from an ethical standpoint. Also, are there plans to integrate more languages or dialects in the future? Excited to see where this goes—keep up the great work!
Maker
@allen_xu1130 Yes, we are adding more languages and improving our repo to make it easy to use :)
Impressive update! The multilingual support and open-source nature are fantastic. How does the voice cloning compare to other TTS engines? Looking forward to testing this out - seems like a game-changer for accessible voice tech!
Can the tone of voice be completely imitated? That’s great. If you add AI face-changing, you can do a lot of interesting things with Fish Speech.
Loving the idea of real-time voice generation with ultra-low latency. Fish Audio is about to change the game!
Congratulations to the entire team for the amazing launch!
Sounds really powerful! 🎧 The fact that it's now trained on 700k hours and supports 8 languages is a huge plus. Congrats on the open-source release! 🚀
This looks like a great solution for creating natural-sounding speech. I’m curious to see how easy it is to use.
Congrats on the launch of Fish Audio! Really excited about the multilingual support. Quick question—do you plan on adding more languages in the future or improving the voice cloning for certain accents? Would love to see that!
Congratulations on the launch of Fish Audio 1.4! 🎉 Making this powerful multilingual voice technology open-source is a game-changer for developers and researchers looking for ultra-fast TTS and instant voice cloning capabilities. Looking forward to seeing the incredible projects that will emerge from this!
Fish Audio 1.4 is a major leap forward in voice technology with 700k hours of multilingual training data. The ultra-fast TTS and instant voice cloning features are game-changers, making cutting-edge voice tech more accessible than ever. Can't wait to see the innovative projects this enables!