Zonos Text-to-Speech
A leading open-weight text-to-speech model trained on more than 200k hours of varied multilingual speech, delivering expressiveness and quality on par with—or even surpassing—top TTS providers.
Key Features:
- Zero-shot TTS with voice cloning
- Multilingual support (EN, JP, CN, FR, DE)
- Audio quality and emotion control
- Real-time generation (2x speed on RTX 4090)
Try Zonos Online
Experience the power of Zonos text-to-speech directly in your browser. No installation required.
Features
What makes Zonos special
Zonos is a leading open-weight text-to-speech model that combines high quality, flexibility, and ease of use.
Zero-shot TTS with voice cloning
Input desired text and a 10-30s speaker sample to generate high quality TTS output
Audio prefix inputs
Add text plus an audio prefix for even richer speaker matching. Audio prefixes can be used to elicit behaviours such as whispering
Multilingual support
Zonos-v0.1 supports English, Japanese, Chinese, French, and German
Audio quality and emotion control
Fine-grained control of many aspects including speaking rate, pitch, maximum frequency, audio quality, and various emotions
Fast generation
Our model runs with a real-time factor of ~2x on an RTX 4090 (generates 2 seconds of audio per 1 second of compute time)
Simple installation and deployment
Zonos comes packaged with an easy to use gradio interface and can be installed and deployed simply using docker
What People Are Saying
See what the community thinks about Zonos
Wow that's very impressive
— Paul Couvert (@itsPaulAi) February 10, 2025
Zonos is a 100% open source AI model that can clone any voice 🤯
You can basically run it anywhere as it's only 1.6B parameters.
Link belowpic.twitter.com/yztyL46NvN
自分の声質でテキストから音声を瞬時に複製できる「Zonos-v0.1」が公開。話す速度や喜怒哀楽を含めた感情も調整可能。つまり音声クローンを創り上げることが出来る。オーディオファイルで学習するので、結論自分の声質以外も学習可能なので、悪用厳禁。詳細はリプ欄へ。 pic.twitter.com/VixYXJXrAG
— みるぼん@スモビジ (@milbon_) February 11, 2025
新发布的最强开源语音模型 Zonos
— 歸藏(guizang.ai) (@op7418) February 11, 2025
语音生成质量非常高，而且这次有中文
- 两种1.6B 模型，transformer 和 SSM
- 用5到30秒的语音进行高保真语音克隆
- 可以调节速度，音高，音频质量和情绪
- 添加文本和音频前缀，实现更丰富的说话人匹配效果
-在 RTX 4090 显卡上运行时，实时率约为 2 倍 pic.twitter.com/hF6qa9JqKW
This is not being talked about enough
— AP (@angrypenguinPNG) February 14, 2025
Zonos is a new open-source voice AI model that clones any voice in under 10 seconds.
Here is how I made a voice clone of @mreflow ! pic.twitter.com/SDEkAH1HOJ
Run ZONOS Locally
— cocktail peanut (@cocktailpeanut) February 15, 2025
ZONOS, the new SOTA Open Source Voice Cloning TTS, is here.
I've managed to write a 1-click launcher for Zonos that works on Mac, Windows, and Linux (ALL platforms!)
Here's me cloning Peter Griffin's voice on my Mac. https://t.co/XLB6NEU8gE pic.twitter.com/nQ1cVtXE4P
What the heck, this is an AI-generated voice and open source too under Apache 2.0 license. Damn! 🔥🔥
— AshutoshShrivastava (@ai_for_success) Febr /uary 10, 2025
Zonos Beta is a new open-source highly expressive TTS model with high-fidelity voice cloning from @ZyphraAI
Try it here 👇pic.twitter.com/GGn877CzGp
