Skip to content
CallMissed
Book a demo

Explore CallMissed

Back to all models
Text to Speechindian-languages

Gnani Timbre v3

by Gnani · Released 2026

Gnani.ai's text-to-speech model with 12 voices, each speaking all 10 supported Indian languages plus Hinglish. Expression is steered with inline audio tags such as <laugh>, <sigh> and <whispers> written directly in the text, with streaming output for live voice agents.

Text to Speech

Gnani Timbre v3

Powered by Gnani · Gnani Timbre neural TTS

Context Window

N/A

Parameters

Undisclosed

Max Output

N/A

Category

Text to Speech

Overview

Gnani Timbre v3 is Gnani.ai's text-to-speech model, offering 12 voices: seven feminine (Shivani, Lakshmi, Kalpana, Chetana, Vasudha, Riyaa, Tarika) and five masculine (Avinash, Shubham, Prakash, Devansh, Yashraj). Every voice speaks all 10 supported languages (Hindi, English, Tamil, Telugu, Kannada, Malayalam, Marathi, Punjabi, Bengali and Gujarati), along with automatic language detection and Hinglish code-switching through the hi-en code.

Each voice has a described role. Shivani is a warm, patient, empathy-first voice for inbound customer support and is the default. Kalpana, Chetana and Tarika cover customer care, healthcare, telecom, banking and insurance; Riyaa and Shubham are energetic voices for outbound sales and lead qualification; Avinash and Prakash are steady, deliberate voices for collections, payment recovery and BFSI; Devansh is a young conversational voice for sales and fintech; Lakshmi, Vasudha and Yashraj are narrative voices for storytelling, e-learning and IVR narration.

Expression comes from inline audio tags that the text itself carries: <laugh>, <giggle>, <sigh>, <exhales>, <sniffles>, <yawns>, <coughs>, <clears throat>, <whispers>, <shouting>, <excited>, <slow> and <pauses>. Tags are case-sensitive, written in angle brackets, and a tag applies from where it sits to the end of the input. SSML is not supported. Output streams, and the REST speech endpoint accepts a speed between 0.85 and 1.15. Pricing is $0.27 per 10K characters.

Pricing

MetricPrice
Price /10K chars₹27.0000

1 credit = ₹1 ≈ US$0.0104 (US$1 = ₹96). Transparent per-call pricing — you pay only for what you call, with no seat fees or minimums.

Key Highlights

  • 12 voices (7 feminine, 5 masculine), each with a described role
  • Every voice speaks all 10 Indian languages, plus auto and Hinglish
  • Inline audio tags: laugh, sigh, whispers, excited, pauses and more
  • Streaming output for real-time voice agents

Technical Details

  • 12 voices: Shivani (default), Lakshmi, Kalpana, Chetana, Vasudha, Riyaa, Tarika, Avinash, Shubham, Prakash, Devansh, Yashraj
  • Languages: Hindi, English, Tamil, Telugu, Kannada, Malayalam, Marathi, Punjabi, Bengali, Gujarati, plus auto-detect and Hinglish (hi-en)
  • Audio tags (case-sensitive, angle brackets): <laugh> <giggle> <sigh> <exhales> <sniffles> <yawns> <coughs> <clears throat> <whispers> <shouting> <excited> <slow> <pauses>
  • A tag applies from where it sits to the end of the input; no SSML
  • Streaming output supported
  • speed 0.85 to 1.15 on the REST /v1/audio/speech API only

Strengths

  • Every voice covers all 10 Indian languages, so one voice can serve a whole customer base
  • Voices described by role: support, sales, collections, healthcare, narration
  • Expression controlled in the text with audio tags
  • Streaming output for real-time voice agents

Limitations

  • No SSML: use audio tags instead
  • Audio tags are case-sensitive and apply to the end of the input, not to a single phrase
  • speed is available on the REST speech API only

Use Cases

Inbound customer supportOutbound sales and lead qualificationCollections and BFSIE-learning and IVR narration

API Example

curl https://api.callmissed.com/v1/audio/speech \
  -H "Authorization: Bearer cm_YOUR_KEY" \
  -d '{"model": "gnani-timbre-v3.0", "input": "Namaste, aapka order confirm ho gaya hai. <excited> Aaj hi delivery hogi!", "voice": "Shivani"}' \
  --output speech.mp3

Endpoint: POST /v1/audio/speech · Model ID: gnani-timbre-v3.0

Try Gnani Timbre v3 now

Get 1000 free API credits on signup. No credit card required.

Start free Book a DemoRead docs