Skip to content
CallMissed
Book a demo

Explore CallMissed

Back to all models
Text to Speechvoice-agentsmultilingualindian-languages

TTS-2

by Inworld · Released Undisclosed

Inworld's newest text-to-speech model. Every voice speaks 94 languages including Hindi, Marathi, Bengali, Tamil, Telugu, Gujarati, Kannada, Malayalam, Punjabi, Urdu and Odia. 14 voices. Voice sessions only. $0.2604 per 10K characters.

Text to Speech

TTS-2

Powered by Inworld · Streaming neural TTS (Inworld TTS-2)

Context Window

N/A

Parameters

Undisclosed

Max Output

N/A

Category

Text to Speech

Overview

TTS-2 is Inworld's newest text-to-speech model. On CallMissed it runs inside voice agents and voice sessions only. Every one of its 14 voices speaks every supported language: 94 base language codes, including Hindi, Marathi, Bengali, Tamil, Telugu, Gujarati, Kannada, Malayalam, Punjabi, Urdu and Odia.

The default voice is riya, a Hindi voice. Inworld publishes gender and a description for only some voices, so the voice list shows those where it has them. Speaking rate is adjustable per agent.

It performs delivery directions (for example "gently warm"), non-verbal sounds (laugh, sigh, breathe, clear throat, cough) and SSML pauses; a voice agent's prompt teaches them automatically. POST /v1/audio/speech returns an error for it. inworld-tts-2-flash is the faster, lower-cost variant with the same voices and languages.

Pricing

MetricPrice
Price /10K chars₹25.0000

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

  • 94 languages — every voice speaks all of them
  • Covers Hindi plus ten more Indian languages
  • 14 voices, 8 of them Hindi; default riya
  • Voice sessions only — not available on the speech endpoint

Benchmarks

BenchmarkScore
Languages94
Voices14
Price$0.2604
DeploymentVoice sessions only

Technical Details

  • 94 base language codes; every voice speaks every code
  • 14 voices; gender and description are listed only where Inworld publishes them
  • Speaking rate adjustable per agent
  • Delivery directions, non-verbal sounds and SSML pauses
  • Voice sessions only — POST /v1/audio/speech returns 400 for this model

Strengths

  • Broadest Indian-language coverage of the voice-session TTS models
  • Every voice works in every language
  • Lower price than Cartesia Sonic 3.6

Limitations

  • Voice descriptions and gender are missing for some voices
  • Voice sessions only — not available on the speech endpoint

Use Cases

Indian-language voice agentsMultilingual contact centresRegional-language outbound calling

API Example

curl https://api.callmissed.com/v1/voice/sessions \
  -H "Authorization: Bearer cm_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "tts_model": "inworld-tts-2",
    "voice": "riya",
    "language": "hi"
  }'

Endpoint: POST /v1/voice/sessions (voice sessions only) · Model ID: inworld-tts-2

Try TTS-2 now

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

Start free Book a DemoRead docs