详细内容
newRead the Octave 2 announcement blog
The world's most realistic & expressive voice AI
Voice AI models powered by emotional intelligence for creators, developers, and enterprises. Create audio books, podcasts, conversational agents and more
Text-to-SpeechVoice DesignVoice CloningConversational AI
Text-to-speech
Heavens no! Please, do what you will with my actuators but, if I may say so, I would much prefer that you leave my motherboard intact!
Narrate a Story
Conversational
134/500
Language
English
Voice
Fastidious Robo-Butler
Play audio
Get started with the most expressive text-to-speech
Trusted by teams at
A text-to-speech system that understands what it's saying
Octave (Omni-capable text and voice engine) isn't a traditional TTS model. It’s a voice-based LLM. That means it understands what words mean in context, so it can predict emotions, cadence, and more.
Announcing Octave 2 from Hume AI on Vimeo
Playing in picture-in-picture
Share
Play
00:00
01:13
Settings
QualityAuto
SpeedNormal
Picture-in-PictureFullscreen
For creators
Generate life-like AI audio for your content creation needs
Audiobooks
Create high-quality mutli-character audiobooks. Upload your PDF, select your characters, direct delivery and publish.
Video voiceovers
Choose the perfect voice for your video or clone your own voice. Then generate high-quality voiceovers for ads, shorts, or feature-length films.
Podcasts
Create multi-speaker podcasts that sound like real, studio quality dialogue. Select your voices, generate audio, and download.
'TwasthenightbeforeChristmas,whenallthroughthehouse.Notacreaturewasstirring,notevenamouse.Thestockingswerehungbythechimneywithcare.InhopesthatStNicholassoonwouldbethere.
Play audio
I'm,like,stunned.It'stotallyinsanetome,honestly.Like,can'ttheyjustwaitoneday?
Play audio
Sohe,uh,hegoesdownthiscrazyrathole.Imean,pictureit.It's1954.He'sstandingbarefootonthebackofallamaandheyellsatthetopofhislungs-
Play audio
Audiobooks
Create high-quality mutli-character audiobooks. Upload your PDF, select your characters, direct delivery and publish.
'TwasthenightbeforeChristmas,whenallthroughthehouse.Notacreaturewasstirring,notevenamouse.Thestockingswerehungbythechimneywithcare.InhopesthatStNicholassoonwouldbethere.
Play audio
Video voiceovers
Choose the perfect voice for your video or clone your own voice. Then generate high-quality voiceovers for ads, shorts, or feature-length films.
I'm,like,stunned.It'stotallyinsanetome,honestly.Like,can'ttheyjustwaitoneday?
Play audio
Podcasts
Create multi-speaker podcasts that sound like real, studio quality dialogue. Select your voices, generate audio, and download.
Sohe,uh,hegoesdownthiscrazyrathole.Imean,pictureit.It's1954.He'sstandingbarefootonthebackofallamaandheyellsatthetopofhislungs-
Play audio
For developers
Build the most realistic, high quality audio models into your products with our APIs and SDKs
Text to speech API
Octave (Omni Capable Text and Voice Engine) is a new kind of voice AI that understands what words mean in context, unlocking humanlike expressiveness and nuance for any voice. With <200ms latency, Octave 2 is available in 11+ languages including:
English
Japanese
Korean
Spanish
French
Portuguese
Italian
German
Russian
Hindi
Arabic
Speech to speech API
EVI (Empathic Voice Interface) is a speech-to-speech foundation model. The same intelligence understands and generates language and speech, enabling it to sound human. You can seamlessly merge EVI's low-latency responses with any language model, including:
Hume
Claude Sonnet 4.5
Grok 4 Fast
GPT 5
+20 more
Tools for Every Platform
Easy to get started with our cross-language development suite:
Python SDK
Typescript SDK
Swift SDK
React SDK
.NET SDK
For enterprise
Deliver new experiences and save costs for your enterprise
AI characters
Power your game character or AI companion with Hume's text to speech or speech to speech API. Deliver expressive and reliable interactions at a cost that makes sense for your application.
Content creation tools
Integrate the most realistic AI voices into your media creation platform. Let your users access hundreds of high-quality voices instantly in over 11 languages for their audiobooks, podcasts, or videos.
Phone calls
Give your AI customer support agent or sales rep the most realistic voice at low latency.
Niantic Spatial x Hume AI: Creating Interactive & Spatially Aware AI Companions
AI characters
Power your game character or AI companion with Hume's text to speech or speech to speech API. Deliver expressive and reliable interactions at a cost that makes sense for your application.
Niantic Spatial x Hume AI: Creating Interactive & Spatially Aware AI Companions
Content creation tools
Integrate the most realistic AI voices into your media creation platform. Let your users access hundreds of high-quality voices instantly in over 11 languages for their audiobooks, podcasts, or videos.
Creating Podcasts at Scale with Inception Point
Phone calls
Give your AI customer support agent or sales rep the most realistic voice at low latency.
Voice AI for Customer Support
We research foundation models and how to align them with human well-being
Octave 2: next-generation multilingual voice AI\
\
October 1, 2025
What is semantic space theory?\
\
February 21, 2024
EVI 3 · Hume AI
The world’s most realistic and instructible speech-to-speech foundation model
The next iteration of the Empathic Voice Interface is here.
Design a voice
Create a custom voice based on a conversation or a text description.
Clone your voice
Upload an audio file to create a clone of your voice.
Podcast Host
Spanish Teacher
Outgoing Friend
Fastidious Robo-Butler
Experiencing high load, please wait…
Staging
Show React Query
1,232 × 624
XSSMMDLGXL2XL
Create experiences with the most empathic AI voices today
Prompt the first LLM for text-to-speech to create new voices, instruct emotions, and more
Get started for free todayGet started for free
Sign up for our newsletter to hear our latest scientific and product updates
Signup Type