Smallest.ai launches Lightning – World's fastest text-to-speech model
- Lightning generates upto 10 seconds of audio in just 100ms
- Generates hyper-realistic audio in English and Hindi
- Drastically reduces the cost for all voicebot builders
BENGALURU, India, Nov. 5, 2024 /PRNewswire/ — Smallest.ai, a leading developer of multi-modal AI foundation models headquartered in San Francisco, California, announces the launch of Lightning the world’s fastest real-time text to speech model that can generate up to 10 seconds of audio in just 100ms. This enables voicebot providers globally to build ultra-realistic bots with sub-second latency, whilst drastically simplifying the integration process.
What makes Lightning revolutionary is also the fact that it costs drastically lower than Western competitors with prices starting at 0.02 USD/min (1.6 Rs/min). This enables population-scale use cases for voicebots at sub 1 Rs/min cost.
Traditionally, real-time text-to-speech models require streaming, which opens up a websocket connection, increasing the computational load on servers and making it difficult and more expensive to scale voicebots. Lightning enables bot providers to scale faster while drastically cutting down on API costs by making it possible to get audio via a simple REST API in around 100ms.
Lightning currently supports multiple English and Hindi accents. Smallest.ai plans on adding other Indian, European, and Asian languages as well in the next couple of months.
Voicebot platforms with early access to Lightning cut their per minute cost by 8x along with a drastic improvement in voice quality.
Whilst Lightning is intended for real-time use-cases, it can also be used to create audiobooks, and voiceovers for reels on Instagram, YouTube, and many more social media platforms. For non-developers, Lightning is available on the Waves Speech platform where support for voice cloning, accent conversion, and other features are available in beta mode.
When asked about it’s mission, Sudarshan Kamath, Founder of Smallest.ai said, “Why are 1B humans not speaking to AI voices everyday despite incredible advancements in Voice AI? This is the problem we are trying to solve.”
About Smallest.ai:
Started by IIT graduates Sudarshan Kamath and Akshat Mandloi, Smallest.ai is backed by marquee investors like 3one4 Capital, Better Capital and Upsparks Capital.
Smallest.ai’s platform is available on Waves and the team actively engages with their customers on discord which is one of the world’s largest speech AI communities.
Photo: https://mma.prnewswire.com/media/2548149/Smallest_ai_launches_Lightning.jpg
View original content to download multimedia:https://www.prnewswire.com/in/news-releases/smallestai-launches-lightning—worlds-fastest-text-to-speech-model-302295555.html
