Chariot Launches Text-to-Speech Model for 11 Indian Languages
The Delhi-based AI lab's Chariot-1-TTS-Multilingual offers more than 50 voices, code-mixing and voice cloning.
Key Takeaways
- Chariot launched Chariot-1-TTS-Multilingual on 12 November 2026.
- The model supports 11 Indian languages and English, with more than 50 voices.
- It can mix two languages in one sentence and clone a voice from reference audio.
- It handles up to 1,000 characters in a single request.
- Chariot is one of the IndiaAI Mission's foundation model teams.
Table of Contents 5 sections
Chariot, a Delhi-based AI research lab, launched Chariot-1-TTS-Multilingual on 12 November 2026. It is a text-to-speech model that turns written text into spoken audio in 11 Indian languages and English.
The model offers more than 50 voices and can mix any two of its supported languages within one sentence. It can also clone a voice from reference audio and handles up to 1,000 characters in a single request.
Chariot-1-TTS-Multilingual at a glance
| Detail | Value |
|---|---|
| Company | Chariot |
| Model | Chariot-1-TTS-Multilingual |
| Launched | 12 November 2026 |
| Languages | 11 Indian languages and English |
| Voices | More than 50 |
| Code-mixing | Any two supported languages in one sentence |
| Voice cloning | From reference audio |
| Text per request | Up to 1,000 characters |
| Headquarters | Delhi |
How Chariot's model decides how to read numbers and symbols
Chariot says the model uses context to decide how written expressions should be spoken, instead of relying on separate text normalisation rules. One example the company gives is "B+", which it reads as "B positive" when it is a blood group. The model also reads currency amounts, dates and abbreviations based on context.
Chariot said the model is meant for uses such as voice agents, customer service and learning platforms. Developers can pick voices from its catalogue or create custom voices from reference recordings.
Who is behind Chariot and its speech models
Chariot was founded by Parth Sarthi and Suvrat Bhooshan. The company is one of the teams selected under the IndiaAI Mission to build foundation models.
Before this launch, Chariot worked on voice projects for brands such as Zomato, Coca-Cola and Cadbury, and on projects with Doordarshan.
| When | What happened |
|---|---|
| Earlier | Voice projects for Zomato, Coca-Cola, Cadbury and Doordarshan |
| October 2026 | Launched Chariot-1-ASR Realtime, a speech recognition model for 10 Indian languages and Indian English |
| 12 November 2026 | Launched Chariot-1-TTS-Multilingual for 11 Indian languages and English |
Where Chariot stands after its text-to-speech launch
With this launch, Chariot's Chariot-1 family covers both speech recognition and text-to-speech. The new model covers 11 Indian languages and English with more than 50 voices. Chariot is one of the teams selected under the IndiaAI Mission to build foundation models.
More news
Frequently Asked Questions
What is Chariot?
What did Chariot launch?
How many languages does Chariot's new model support?
How many voices does Chariot's model offer?
How does Chariot's model handle mixed-language sentences?
How does Chariot's voice cloning work?
How much text can Chariot's model read in one request?
How does Chariot's model read numbers and symbols?
Who founded Chariot?
Is Chariot part of the IndiaAI Mission?
What other model has Chariot launched?
Who is Chariot's new model meant for?
More in AI
OpenAI Starts Local API Processing in India With GPT-6 Luna
OpenAI now lets eligible API customers process supported GPT-6 Luna requests inside India.
IntellectAI Launches AI Portfolio Tool for Insurers
IntellectAI has launched Xponent Portfolio Navigator, an AI portfolio steering tool for insurers.
Wipro Partners With Labelbox to Take AI Agents to Production
Wipro and Labelbox will help enterprises move AI agents from pilot projects into production.