The first LLM for text-to-speech
Prompt to create AI voices, change emotions, and more
Trusted By
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.
Create any voice you can imagine with Octave Voice Design
"sarcastic medieval peasant"
Full prompt: "The speaker is a medieval peasant with a cockney accent, raspy voice, dripping with sarcasm."
"literature professor"
Full prompt: "A retired Black female literature professor who analyzes poetry with precise academic language and references to her own published criticism."
"charming cowboy"
Full prompt: "The speaker is a grizzled old cowboy with a folksy Texan drawl Southern accent, speaking in a charismatic tone with a deep but relaxed vibe."
"sitcom inner monologue"
Full prompt: "The star of a popular sitcom, with frequent inner monologues about her life."
"dungeon master"
Full prompt: "A know-it-all dungeons and dragons dungeon master speaking excitedly with a lisp."
"warm English narrator"
Full prompt: "The speaker is a sophisticated British female narrator with a gentle, warm voice, recounting the ending of a classic romance novel."
"unserious movie trailer guy"
Full prompt: "The speaker is an American, deep middle-aged male film trailer narrator for a film about chickens."
"raspy evil vampire"
Full prompt: "A villainous undead vampire, with a horrifying raspy voice, and a slight Transylvanian accent."
"reminiscing"
Full prompt: "A middle-aged African American man, reminiscing with a slightly gravelly voice and a tone of hard-earned wisdom."
"nature documentary narrator"
Full prompt: "The speaker is a distinguished British narrator, whose voice carries a deep sense of wisdom and curiosity."
"Texan fishing guru"
Prompt: "The speaker has a booming, charismatic radio voice, like a Texan fishing guru with a hint of gravel and an infectious laugh, perfect for reeling in listeners to 'Big Dicky's live fishing frenzy.'"
Generating the best AI voices has never been easier
In a blind comparison study with over 100 human raters, Octave’s outputs were favored over outputs from ElevenLabs Voice Design in terms of audio quality, naturalness, and how well speech generations matched descriptions of the desired voice, across 120 diverse prompts.
The first AI voice generator that can take nuanced Acting Instructions
"whispering, hushed"
Here, we combine the text "Are you serious?" with the prompt "whispering, hushed."
“angry, furious"
With speaker and text held constant, we change the prompt to "angry, furious."
"calm, serene"
With speaker and text held constant, we change the prompt to "calm, serene."
“disgusted, disdainful”
With speaker and text held constant, we change the prompt to "disgusted, disdainful."
"pained, shocked"
With speaker and text held constant, we change the prompt to "pained, shocked."
Any emotion or speaking style, on command
Octave is the first TTS system that can take natural language instructions to change emotional delivery and speaking style. Give directions like "sound sarcastic" or "whisper fearfully." For the first time, creators have total control.
For creators and developers alike
Octave was built to generate the most expressive AI voices for any content: podcasts, voiceovers, audiobooks, and more. With our API, you can bring it to any application.
References
.png?width=1080&height=1080&quality=75&format=webp&fit=inside)
Octave TTS: the first text-to-speech system that understands what it’s saying
Today we’re launching Octave (Omni-capable text and voice engine), the first LLM for text-to-speech. Unlike conventional TTS that merely “reads” words, Octave is a speech-language model that understands what words mean in context, unlocking a new level of expressiveness and nuance—and new AI voice capabilities.
Developer Documentation
Explore our documentation with concise guides, hands-on tutorials, and an in-depth API reference—crafted to support your integration.
Controlling the speed of AI voices
The rise of artificial intelligence has revolutionized many aspects of our lives, and one area where its impact is increasingly felt is in the realm of voice technology. AI voices are now commonly used in virtual assistants, customer service bots, audiobooks, video games, and various accessibility tools. As this technology continues to evolve, a key question emerges: to what extent can we control the speed (speech rate) of these synthetic voices?
00/00