Human listeners ranked Lightning TTS first among real-time models on Voice Arena, ahead of ElevenLabs, Cartesia and Sarvam.
SAN FRANCISCO, CA, UNITED STATES, July 24, 2026 /EINPresswire.com/ — Smallest.ai announced today that its text-to-speech model, Lightning, has been ranked the top real-time voice for Hindi customer support on Voice Arena.
Lightning ranked first among real-time models in the category, ahead of leading providers including ElevenLabs, Cartesia and Sarvam. Voice Arena uses blind comparisons in which listeners hear voices without knowing which models produced them and vote for the one they prefer.
The Hindi customer-support category reflects how Indian enterprises use voice AI across contact centres, collections and customer-support operations in banking, insurance and BPO.
The result builds on a wider pattern. Smallest.ai evaluates Lightning on the factors that shape a real-time voice experience, including naturalness, generation speed and voice quality. These evaluations include internal testing and independent third-party assessments by Voice Arena, Artificial Analysis and Speko.
Many TTS models are developed primarily around English speech patterns. This can affect Hindi pronunciation, rhythm and the natural delivery of mixed Hindi-English text.
Lightning is designed to generate natural Hindi and Hinglish speech in real time. In Smallest.ai’s internal streaming tests, Lightning produced its first audio as fast as 80 milliseconds. The model preserves pronunciation and pacing during code-switching and supports audio formats commonly used in telephony, simplifying integration into customer-calling systems.
Voice agents built on Smallest.ai’s technology have handled more than 150 million calls across enterprise deployments. This production experience informs how the company optimizes and deploys its technology for real telephony environments, beyond controlled evaluations.
Lightning is available through a single API and can be deployed in the cloud or on premises.
Smallest.ai is an AI research lab building real-time voice infrastructure for enterprises. The company develops speech recognition, speech generation and speech-to-speech systems designed to enable natural, scalable AI conversations across customer service, healthcare, financial services and other high-volume communication environments. The company is backed by Sierra Ventures, 3one4 Capital, Better Capital, Upsparks Capital, Schema Ventures, Tiny VC, DeVC, Mission Street Capital and other angel investors.
Devansh Pawan
Smallest AI
social@smallest.ai
Legal Disclaimer:
EIN Presswire provides this news content “as is” without warranty of any kind. We do not accept any responsibility or liability
for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this
article. If you have any complaints or copyright issues related to this article, kindly contact the author above.
Media gallery
