Skip to content
CallMissed
Book a demo

Explore CallMissed

Back to all models
Text to Speechvoice-agentsmultilingualindian-languagesstreaming
Deepgram — deepgram-flux-tts logo

Flux TTS

by Deepgram · Released Undisclosed

Deepgram's conversational text-to-speech model, built for voice agents. 266 voices in 42 languages, including Hindi, Tamil, Telugu, Kannada, Marathi, Gujarati, Bengali and Urdu. Streaming supported. $0.4688 per 10K characters.

Deepgram — deepgram-flux-tts logo
Text to Speech

Flux TTS

Powered by Deepgram · Streaming neural TTS (Deepgram Flux)

Context Window

N/A

Parameters

Undisclosed

Max Output

N/A

Category

Text to Speech

Overview

Flux TTS is Deepgram's newest conversational text-to-speech model, designed for voice agents rather than long-form narration. It has 266 voices in 42 languages: 149 generally available voices in 10 languages (including Hindi) and 117 early-access voices in 32 languages (including Tamil, Telugu, Kannada, Marathi, Gujarati, Bengali and Urdu).

The default English voice is priya, a self-assured and reassuring Indian-English voice. Each voice belongs to one language, so the voice you pick sets the language. Early-access languages read numbers and dates as written, so spell them out. Use it on POST /v1/audio/speech, on voice agents and voice sessions, and in the managed Voice Agent.

Pricing

MetricPrice
Price /10K chars₹45.0000

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

Key Highlights

  • Built for conversational voice agents
  • 266 voices in 42 languages, with Hindi generally available
  • Early-access voices in Tamil, Telugu, Kannada, Marathi, Gujarati, Bengali and Urdu
  • Streaming on POST /v1/audio/speech with pcm, wav, mulaw or alaw

Benchmarks

BenchmarkScore
Languages42
Voices266
Price$0.4688
Input limit2000 characters

Technical Details

  • 42 languages; the voice selects the language
  • Voice list with gender and language from GET /v1/models/deepgram-flux-tts
  • Input up to 2000 characters; without stream: true, up to 400
  • Streaming output: pcm, wav, mulaw or alaw
  • Plain text only: no SSML or inline sound tags

Strengths

  • Hindi plus early-access Tamil, Telugu, Kannada, Marathi, Gujarati, Bengali and Urdu
  • Designed for turn-based voice-agent conversation
  • Available on the speech endpoint and in voice agents

Limitations

  • Early-access languages read numbers and dates as written
  • Without streaming, input is limited to 400 characters
  • Speed control for English voices only

Use Cases

Hindi and Indian-English voice agentsMultilingual customer supportConversational IVR

API Example

curl -N https://api.callmissed.com/v1/audio/speech \
  -H "Authorization: Bearer cm_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepgram-flux-tts",
    "voice": "ajay",
    "input": "Namaste, aap kaise hain?",
    "response_format": "pcm",
    "stream": true
  }' --output speech.pcm

Endpoint: POST /v1/audio/speech · Model ID: deepgram-flux-tts

Try Flux TTS now

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

Start free Book a DemoRead docs