Lightning V3 - Text-to-Speech built for Voice Agents
by•
Introducing Lightning V3 — Smallest AI's most advanced text-to-speech model. With 100ms latency, a 3.89 WVMOS score, and support for English, Hindi, Spanish, Tamil and 15+ languages, V3 was preferred over OpenAI's GPT-4o-mini-TTS by listeners 76.2% of the time.
Get audio output in 44.1 kHz and powers voice assistants, IVR systems, content creation and conversational AI with human-like speech. Instant voice cloning from just 10 seconds of audio. Real-time. Expressive. Enterprise-ready.
Replies
I used the voice clone feature! I am able to use the realistic voices in videos that I create loving the experience.
FuseBase
Huge congrats team!! 🚀
voice AI that actually sounds human is still rare tbh
gonna test this later today, been looking for smth like this for a side project
How is it different from ElevenLabs?
Documentation.AI
This is actually very cool. 100ms latency coupled with decent prosody is kinda the holy grail for voice agents.
ConnectMachine
76.2% preference over GPT-4o-mini-TTS is impressive. Would love to know how big the test group was and what kind of prompts were used??
Voice cloning from 10 seconds of audio is impressive. We're doing something similar but for writing — Krafl-IO analyzes your LinkedIn posts and clones your writing voice so AI-generated posts sound like you wrote them. Different modality, same core challenge: Capturing someone's unique style. Congrats on #2!