Skip to content
CallMissed
Book a demo

Explore CallMissed

Back to all models
Text to Speechvoice-agentsmultilingualindian-languagesrealtime

Eleven v3 Conversational

by ElevenLabs · Released Undisclosed

Eleven v3 tuned for natural dialogue in conversational agents. 85 languages including 11 Indian languages; audio tags. $0.4167 per 10K characters.

Text to Speech

Eleven v3 Conversational

Powered by ElevenLabs · Neural TTS (ElevenLabs Eleven v3 Conversational)

Context Window

N/A

Parameters

Undisclosed

Max Output

N/A

Category

Text to Speech

Overview

Eleven v3 Conversational is the realtime variant of Eleven v3: the same expressive delivery and audio tags, optimised for natural back-and-forth in voice agents (ElevenLabs quotes about 165 ms to first audio). It speaks 85 languages, including Hindi, Bengali, Tamil, Telugu, Marathi, Gujarati, Kannada, Malayalam, Punjabi, Odia and Urdu.

Half the price of Eleven v3 per character. Available on POST /v1/audio/speech and as a voice-agent voice; shares the ElevenLabs voice library.

Pricing

MetricPrice
Price /10K chars₹40.0000

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

  • Expressive conversational agents
  • 85 languages, one shared ElevenLabs voice library (21 premade voices or any ElevenLabs voice ID)
  • On POST /v1/audio/speech and as a voice-agent voice (tts_model)
  • Speed 0.7–1.2; mp3, wav, pcm, opus, mulaw and alaw output

Benchmarks

BenchmarkScore
Languages85
Voices21+
Price$0.4167
Latency~0.55 s to first audio in our tests

Technical Details

  • Model ID is ElevenLabs' own (eleven_v3_conversational)
  • 85 languages; language takes a base ISO code such as hi
  • Voice: a premade name or any 20-character ElevenLabs voice ID, including cloned voices
  • Input up to 4096 characters; no SSML; audio tags such as [laughs] are performed
  • Paid plans only

Strengths

  • Eleven v3 tuned for natural dialogue in conversational agents
  • Cloned and library voices from your ElevenLabs account work as-is
  • Available on the speech endpoint and in voice agents

Limitations

  • Premade voices carry English accents; the language is enforced by the language parameter
  • No SSML; streaming is not supported on the speech endpoint
  • Paid plans only

Use Cases

Expressive conversational agentsCustomer support and sales voice agentsSpoken notifications and IVR prompts

API Example

curl https://api.callmissed.com/v1/audio/speech \
  -H "Authorization: Bearer cm_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "eleven_v3_conversational",
    "voice": "sarah",
    "language": "hi",
    "input": "Namaste, main aapki kaise madad kar sakti hoon?"
  }' --output speech.mp3

Endpoint: POST /v1/audio/speech · Model ID: eleven_v3_conversational

Try Eleven v3 Conversational now

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

Start free Book a DemoRead docs