Today Hume is introducing Real World VoiceEQ: a benchmark for measuring the human quality of voice AI across 40+ proprietary and open-source models, 15+ dimensions, and 60+ metrics—developed from more than 1 million individual human ratings.
One performance, infinite voices.
Voice Conversion is now live on Hume’s creator studio and API!
Generate the same pacing, pronunciation, and intonation with one recording across any voice you choose.
Hear it for yourself ⬇️
Why are phone numbers so hard for voice AI?📞🤖
Did you know that reading out something as simple as 555-010-2314 is a weirdly difficult task for TTS?
In a recent eval we ran on all the major TTS models, Hume’s Octave 2 comes out at the top with the lowest error rates for
Hume's TTS is now available for free in Adobe Express.
Use the Hume plugin to instantly create voices and generate samples all within the Adobe Express app.
Bring your videos, presentations, or graphics to life with the world’s most realistic voice AI!
Today, @NianticSpatial released an update to their AR companion, Dot, at Snap's Lens Fest, with new voice capabilities powered by Hume AI.
Dot's new interactive dialogue capabilities allow the AI companion to guide users through physical spaces, offering contextual information
We'll be releasing more evaluations, more languages, and access to voice conversion and phoneme editing soon. In the meantime, we’re excited to see what you create!
Read more about product & research in our blog ↓ https://www.hume.ai/blog/octave-2-launch
Introducing Octave 2: our next-generation multilingual text-to-speech model
What’s new:
- Fluent in 11+ languages
- 40% faster (<200ms latency) & 50% cheaper than Octave 1
- Multi-speaker conversation
- More reliable pronunciation
- New voice conversion & phoneme editing
We’ll be releasing Spanish and German officially next week, with Portuguese, Japanese, French and a lot more releasing soon.
Sign up for free to stay updated at http://platform.hume.ai
EVI 3 is perfect for AI companions, interviews, coaching, and learning use cases.
You can also start with Octave, our expressive TTS model, perfect for content generation. Octave TTS has also recently been deployed on LiveKit, Vapi, Vercel, and Twilio to serve a wide range of
Hume’s EVI 3 is now available via API or http://demo.hume.ai. Use our own language generation or any of the external LLMs like Claude 4, Gemini 2.5, and Kimi K2, which can be merged seamlessly with EVI 3’s natively generated quick responses.
2024: Voice Cloning
2025: What about personality cloning?
Hume’s voice AI can now not only mimic your voice but also speaking style and language.
It’s now available via our TTS and new speech-to-speech model, EVI 3, which is also launching today.
Check our new MCP server! Now you can use Hume Text-to-Speech directly from an AI chat, using clients like Cursor or Claude Desktop.
See @twitchard use @Cursor to create a voiceover in less than a minute.
EVI 3 can transform its delivery implicitly or on command—it stammers anxiously, debates enthusiastically, and whispers intimately.
The model can also generate any voice and personality from your prompt in under a second. Speak to a “raspy Australian history buff,” “sassy
Meet EVI 3, another step toward general voice intelligence.
EVI 3 is a speech-language model that can understand and generate any human voice, not just a handful of speakers. With this broader voice intelligence comes greater expressiveness and a deeper understanding of tune,
Octave TTS is now 𝑓𝑎𝑠𝑡
Introducing Octave Instant Mode.
Now, the highest quality TTS on the market runs in under 250ms…
4x faster than the second best model...
while retaining all the nuance, emotion, and personality that sets it apart.
We just updated our text-to-speech model, Octave, with even better audio quality and naturalness.
Now, it's best-in-class for even the most basic TTS applications.
In a blind rating study comparing Hume's updated Octave TTS to Elevenlabs and 10 other TTS models for simple,
OpenAI enters the Expressive TTS Arena 🥊🤖🥊
Now hosted on @huggingface, this arena is a new way to evaluate voice AI systems with natural language instructions + richer text.
Compare Hume's TTS against ElevenLabs and OpenAI and see if you agree with the leaderboard results!
