Skip to content

Explore CallMissed

Speech to Textvoice-agentsstreamingmultilingualindian-languages

Scribe v2 Realtime

by ElevenLabs · Released Undisclosed

ElevenLabs' real-time speech recognition for voice agents: 100 languages with automatic detection, word-level timestamps. Voice sessions only. $0.4063 per audio hour.

Speech to Text

Scribe v2 Realtime

Powered by ElevenLabs · Streaming ASR (ElevenLabs Scribe v2)

Context Window

N/A

Parameters

Undisclosed

Max Output

N/A

Category

Speech to Text

Overview

Scribe v2 Realtime is ElevenLabs' live speech recognizer (about 150 ms to a committed transcript upstream) with the same 100 languages as Scribe v2 and automatic language detection, so a multilingual agent can follow the caller. It has no turn detection of its own, so the agent's standard turn detection applies.

On CallMissed it runs inside voice agents and voice sessions only (stt_model). For recorded audio use scribe_v2.

Pricing

MetricPrice
Price /hour₹39.0000

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

  • 100 languages including 13 Indian languages, automatic language detection
  • Streaming for voice agents — not available for file transcription
  • ElevenLabs' real-time speech recognition for voice agents

Benchmarks

BenchmarkScore
Languages100
Price$0.4063
DeploymentStreaming only

Technical Details

  • Model ID is ElevenLabs' own (scribe_v2_realtime)
  • language takes a base ISO code such as hi; leave it out to detect the language
  • Selected per agent or voice session via stt_model; no model-native turn detection
  • Paid plans only

Strengths

  • 100 languages with automatic detection
  • ElevenLabs' real-time speech recognition for voice agents: 100 languages with automatic detection, word-level timestamps

Limitations

  • Streaming only — not available for file transcription
  • Paid plans only

Use Cases

Multilingual voice agentsLive call transcriptionIndian-language transcription

API Example

curl https://api.callmissed.com/v1/voice/sessions \
  -H "Authorization: Bearer cm_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "stt_model": "scribe_v2_realtime",
    "language": "hi"
  }'

Endpoint: POST /v1/voice/sessions (streaming only) · Model ID: scribe_v2_realtime

Try Scribe v2 Realtime now

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