Loading video...
Video Failed to Load
Introducing Eleven v3 (alpha) - the most expressive Text to Speech model ever. Supporting 70+ languages, multi-speaker dialogue, and audio tags such as [excited], [sighs], [laughing], and [whispers]. Now in public alpha and 80% off in June.
1,956,373 views • 1 year ago •via X (Twitter)
11 Comments
This is a research preview. It requires more prompt engineering than previous models - but the generations are breathtaking. We’ll continue fine-tuning to improve reliability and control.
The new architecture of Eleven v3 deeply understands text - delivering much greater expressiveness. And now you can guide generations more directly using audio tags: - Emotions [sad] [angry] [happily] - Delivery direction [whispers] [shouts] - Non-verbal reactions [laughs] [clears throat] [sighs]
Generate multi-speaker dialogue that sounds like a real conversation. Eleven v3 handles interruptions, shifts in tone, and emotional cues based on conversational context.
Public API for Eleven v3 (alpha) is coming soon. For early access, please contact sales. We are working on the real-time version of v3. For real-time and conversational use cases, we recommend staying with v2.5 Turbo or Flash for now.
Built for creators and developers building media tools. If you’re working on videos, audiobooks, or media tools - v3 unlocks a new level of expressiveness. Learn how to get the most out of it with our prompting guide:
Eleven v3 (alpha) is available now: It’s 80% off during June. Try it out - and share your best generations with us.
Introducing Vehrbal, the AI that converts audio into SOAP notes! Say goodbye to wasted time and hello to effortless note-taking. Experience the power of fast, simple, and efficient with Vehrbal today.
Congratulations my friends, such incredible work and so cool to see this finally out ❤️
ElevenLabd and @cursor_ai are my fav companies, thanks for all you are doing for us
The amount of value we have just unleashed to the world is unreal. Welcome to the model family v3.
[applause] Major congratulations to the research team on continuing to lead the way in AI Audio - this model can produce some truly magical outputs
![Eleven v3 (alpha) is the most expressive Text to Speech model. v3 introduces: • Multi-speaker dialogue with contextual awareness • Support for 70+ languages, up from 33 in v2 • Audio tags such as [excited], [sighs], [laughing], and [whispers]](https://image.24vids.com/tw-1933188969279500459/amplify_video_thumb/1933183425982214145/img/cs5jLCKLDanwBQ5V.jpg)
![We are launching the Eleven v3 (alpha) API. Built for async use cases, Eleven v3 (alpha) delivers the most expressive Text to Speech model: - Dialogue mode, unlimited amount of speakers - 70+ languages - Enhanced voice and emotional control with [audio tags]](https://image.24vids.com/tw-1958219938881257525/amplify_video_thumb/1958217209127501824/img/VG8hXirxkhOEaLl9.jpg)
![Today we launched Gemini 3.1 Flash TTS, our most expressive and controllable text-to-speech model yet. This launch [excitement] includes audio tags! 🗣🏷 Audio tags [explanatory] are a seamless way to guide vocal style, pace, and delivery using natural language commands embedded directly in your text. Want a different tempo or tone? [amazement] Just tag the audio to steer the AI-speech output! The model supports 70+ languages (24 of which are high-quality evaluated languages, including: Japanese, Hindi, and Arabic). Watch the audio tags in action in the demo below ↓](https://image.24vids.com/tw-2044447560384102592/amplify_video_thumb/2044447449708990464/img/DpiMl_0pWyVPDKFC.jpg)
![Google presents AudioPaLM: A Large Language Model That Can Speak and Listen paper page: introduce AudioPaLM, a large language model for speech understanding and generation. AudioPaLM fuses text-based and speech-based language models, PaLM-2 [Anil et al., 2023] and AudioLM [Borsos et al., 2022], into a unified multimodal architecture that can process and generate text and speech with applications including speech recognition and speech-to-speech translation. AudioPaLM inherits the capability to preserve paralinguistic information such as speaker identity and intonation from AudioLM and the linguistic knowledge present only in text large language models such as PaLM-2. We demonstrate that initializing AudioPaLM with the weights of a text-only large language model improves speech processing, successfully leveraging the larger quantity of text training data used in pretraining to assist with the speech tasks. The resulting model significantly outperforms existing systems for speech translation tasks and has the ability to perform zero-shot speech-to-text translation for many languages for which input/target language combinations were not seen in training. AudioPaLM also demonstrates features of audio language models, such as transferring a voice across languages based on a short spoken prompt.](https://image.24vids.com/tw-1672048076469698561/ext_tw_video_thumb/1672047839378300928/pu/img/MuoOBsAyUd-HAI7D.jpg)