For the complete documentation index, see llms.txt. This page is also available as Markdown.
Page cover

Bodhi Overview

Below is a introduction of Navana.ai's Bodhi ASR Streaming features. Please refer to the corresponding documentation for more details.

🗣️ About Bodhi

Bodhi is Navana.ai's flagship voice agent platform, designed with a clear vision: to make business communication across India scalable, accessible, and intelligent. Powered by our proprietary Speech-to-Text (STT) technology, Bodhi offers highly accurate, multilingual voice agents tailored specifically for Indian enterprises.

With coverage across 10 Indian languages and 40+ dialects, Bodhi today can engage over 900+ million Indian language speakers — making it the largest, most inclusive voice AI platform of its kind. Our technology is built ground-up for India's unique linguistic, cultural, and real-world business environments, delivering an unparalleled voice automation experience for businesses.

🚀 Getting Started

Signing Up

To start using Bodhi APIs, sign up on our website. Upon successful sign up, you will receive an email verification to set up your account.

Accessing Your API Key

Upon successful signup to the Bodhi dashboard, you will receive an API key which allows you to access our ASR (Automated Speech Recognition) service. Your active API key and Customer ID will be displayed in your account dashboard and can be copied for use.

Once your API key is set up, you can start using Bodhi's streaming ASR services with the below listed ASR models.


🎯 Available languages & ASR Models

All models are bilingual (language + English) by default, except Gujarati, Odia, and the dedicated English model.

Language
General Model
Banking Model
Notes

Bengali

bn-general-v2-8khz

bn-banking-v2-8khz

Bilingual with English (code-switching)

English (en-IN)

en-general-v2-8khz

en-banking-v2-8khz

English-only (dedicated model)

Hindi

hi-general-v2-8khz

hi-banking-v2-8khz

Bilingual with English (code-switching)

Hinglish (Hindi–English)

hi-en-general-v2-8khz

hi-en-banking-v2-8khz

Retained for compatibility

Kannada

kn-general-v2-8khz

kn-banking-v2-8khz

Bilingual with English (code-switching)

Malayalam

ml-general-v2-8khz

ml-banking-v2-8khz

Bilingual with English (code-switching)

Marathi

mr-general-v2-8khz

mr-banking-v2-8khz

Bilingual with English (code-switching)

Tamil

ta-general-v2-8khz

ta-banking-v2-8khz

Bilingual with English (code-switching)

Telugu

te-general-v2-8khz

te-banking-v2-8khz

Bilingual with English (code-switching)

Gujarati

gu-general-v2-8khz

gu-banking-v2-8khz

Monolingual

Odia

or-general-v3-8khz

(TBD)

Monolingual

Notes:

  • All models except English, Gujarati, and Odia support seamless code-switching with English. They handle mixed-language input and also perform well on English-only audio.

  • For pure English use cases, we still recommend the dedicated English model, which has recently been updated for improved accuracy.

  • The Hinglish models remain available for backward compatibility, but the Hindi model now includes built-in bilingual support by default.

  • Gujarati and Odia remain monolingual for now; bilingual upgrades for them are in progress and will be reflected here.

Get started with Bodhi by signing up on the Navana.ai's website


🔥 Latency and Concurrency

To measure latency for websockets please refer to Measuring Latency section.

Bodhi’s infrastructure is built to support real-time transcription at scale. If your application has specific concurrency or throughput requirements (e.g., bursty loads, high sustained parallel sessions, or low-latency SLAs), we are happy to work with you to ensure the system meets your needs.

📩 Reach out to support@navanatech.in to discuss custom deployment, concurrency tuning, or dedicated infrastructure options.


Last updated