Chariot, the Delhi-based frontier speech AI lab backed under the IndiaAI Mission, on 12 November 2026 launched Chariot-1-TTS-Multilingual, a text-to-speech model that speaks 11 Indic languages and English. The release gives Indian enterprises building voice agents, helplines and learning apps a homegrown option that the company says leads rival models on Hindi and Indic accuracy.

Chariot is one of the first four teams selected under the IndiaAI Mission to build foundation models for the country. Its launch adds to a fast-growing field of Indian voice AI, from Sarvam’s latest models to new public-sector deployments.

What Chariot-1-TTS-Multilingual can do

• Languages: 11 Indic languages plus English, with American, British and Indian English accents.
• Voices: a catalogue of more than 50 voices, plus instant voice cloning from a few seconds of reference audio.
• Code-mixing: switching between any two supported languages within the same sentence, the way many Indians actually speak.
• Length: up to 1,000 characters of text converted to speech in a single request.
• Context-aware pronunciation: correct reading of payment amounts, grades, blood groups, dates, abbreviations and technical terms without custom normalisation rules.

How Chariot’s model scored on benchmarks

In the company’s own launch comparison, Chariot-1-TTS-Multilingual ranked first on Hindi and Indic accuracy against selected model versions from ElevenLabs, Cartesia, Microsoft, OpenAI, AWS and Sarvam. Chariot reported a character error rate of 7.60% on Hindi and 6.47% across nine other Indic languages in intelligibility tests.

The model also posted the highest English verbalisation accuracy, at 93.78%, on a benchmark of more than 1,000 sentences spanning 40 categories, and led on NISQA, an automated speech quality metric, for plain English. These figures come from Chariot’s launch material and have not yet been independently reproduced.

Why it matters for Indian enterprises

Text-to-speech errors on numbers, dates and mixed-language sentences have long forced Indian banks, telcos and public helplines to write and maintain their own pronunciation rules. A model that handles that context natively can cut the engineering effort behind voice bots for payment reminders, customer service and citizen services.

It also fits the government’s push for indigenous AI. Officials are reportedly weighing wider use of domestic AI models in critical sectors, while other IndiaAI Mission teams have recently released their own foundation models. Voice is a natural early battleground, as seen in projects such as the NHA’s AI voice scribe for public doctors.

What Chariot’s founders said

Co-founder Parth Sarthi said people use numbers, symbols and abbreviations and move between languages in the same sentence, so a speech model must understand what those choices mean. By building that understanding into the model, he said, developers can spend less time writing normalisation rules and more time building voice experiences.

Co-founder Suvrat Bhooshan said voice technology in India must work with the languages people speak and the way they switch between them. Chariot’s ambition, he added, is to build foundation models that understand, generate and reason in speech.

What comes next

Chariot, founded by former Google DeepMind and Meta FAIR researchers, plans to keep expanding its Chariot-1 family across speech recognition and generation. The lab already offers Chariot-1-ASR Realtime for speech recognition, and its official announcement page lists the multilingual TTS model as its latest release.

Frequently asked questions

What is Chariot-1-TTS-Multilingual?

It is a text-to-speech model from Indian AI lab Chariot that converts text into natural speech in 11 Indic languages and English, with more than 50 voices and voice cloning.

Is Chariot part of the IndiaAI Mission?

Yes. Chariot is one of the first four teams selected under the IndiaAI Mission to build foundation models from India.

Be a part of Elets Collaborative Initiatives. Join Us for Upcoming Events and explore business opportunities. Like us on Facebook , connect with us on LinkedIn and follow us on Twitter.

"Exciting news! Elets technomedia is now on WhatsApp Channels Subscribe today by clicking the link and stay updated with the latest insights!" Click here!