Transcriber
ElevenLabs Scribe (STT) logo

ElevenLabs Scribe speech-to-text for Bolna AI voice agents

ElevenLabs Scribe v2 Realtime transcribes callers for a Bolna voice agent in about 150ms, across 90 languages, with word-level timestamps and auto language detection.

stttranscriberelevenlabsspeech-to-textscribemultilingual
At a glance

How ElevenLabs Scribe (STT) fits in the stack

Best for

Turning noisy phone audio into text the agent can reason over.

Use this layer when

Recognition accuracy, language coverage, and latency matter most.

Connects to

Telephony audio upstream and your LLM downstream.

Voice Stack

Speech-to-text is the listening layer

Speech-to-text is the listening layer

This provider converts raw audio into text in real time. It shapes how accurately the agent hears the caller and how natural the back-and-forth feels.

TelephonyPhone Network
→
STTListener
→
LLMReasoning
→
TTSVoice
→
ToolsActions

Last updated

What is ElevenLabs Scribe used for in a Bolna voice agent?

ElevenLabs Scribe turns the caller's speech into text for your Bolna agent, using the scribe_v2_realtime model with about 150ms latency across 90 languages. It returns word-level timestamps, detects the spoken language automatically and decides when a caller has finished with VAD-based endpointing. It is also on Bolna's preferred model list, so it fits inside the flat rate.

How do I set up ElevenLabs Scribe with Bolna?

  1. Open your agent at platform.bolna.ai and go to the Languages tab.
  2. Under Transcription, set Provider to ElevenLabs and Model to scribe_v2_realtime.
  3. Pick the caller's language.
  4. Add up to 50 brand or product terms in Keywords. Weights are dropped.
  5. On the Engine tab, adjust Endpointing if the agent cuts callers off or waits too long. Scribe accepts silence thresholds from 0.3 to 3.0 seconds.

In the API, set provider to elevenlabs and model to scribe_v2_realtime on tools_config.transcriber. See the ElevenLabs transcriber guide in the Bolna docs.

What does ElevenLabs Scribe support on Bolna?

FeatureDetail
Modelscribe_v2_realtime
LatencyAbout 150ms, excluding network overhead
Languages90
Turn detectionVAD-based endpointing, 0.3 to 3.0 second silence threshold
TimestampsWord level, used by Bolna for per-word latency tracking
Language detectionAutomatic, with code-switching detection
Audio on Twiliomulaw at 8kHz
Audio on Exotel and Plivolinear16 at 8kHz
Audio on web callslinear16 at 16kHz
KeywordsUp to 50 terms, no weights, no context field
PricingOn the preferred model list, included in the flat rate

What can you build with ElevenLabs Scribe and Bolna?

One-vendor voice stacks. Use Scribe to hear the caller and ElevenLabs text-to-speech to answer, with a single provider on both ends of the call.

Multilingual agents. With 90 languages and automatic detection, one agent can follow callers who switch language mid-call. Bolna logs the per-word language breakdown for review.

Agents on Bolna's Indian servers. ElevenLabs is one of the transcribers allowed on the Indian server configuration, alongside Azure, Deepgram, Sarvam and Smallest.

Flat-rate campaigns. scribe_v2_realtime is covered by Bolna's flat per-minute rate, so transcription costs stay predictable on high-volume outbound calls. See pricing.

FAQ

ElevenLabs Scribe (STT) on Bolna: frequently asked questions

Yes. Set the transcriber provider to `elevenlabs` and the model to `scribe_v2_realtime`. It streams over a WebSocket and returns text in about 150ms, not counting network overhead.

Yes. `scribe_v2_realtime` is on Bolna's preferred model list for speech-to-text, so it is covered by the flat per-minute rate.

Up to 50 terms. Bolna sends them as keyterms and strips any weights. The Context field is disabled because ElevenLabs has no equivalent.

Yes. With language detection on, Bolna can track when a caller switches language mid-conversation, with a per-word language breakdown.

Browse this layer

Keep exploring the voice stack

Browse

Speech-to-Text

Speech-to-text converts what callers say into text that your LLM can process. Transcription accuracy and latency directly affect how natural a conversation feels. Bolna supports streaming STT providers optimized for telephony audio, including specialized models for Indian languages.

Browse

Telephony

Telephony providers connect your voice agents to the phone network so they can make and receive real calls. Bolna supports managed integrations with major carriers as well as bring-your-own-carrier via SIP trunking, giving you full control over call routing, number provisioning, and cost.

Browse

Large Language Models

The LLM is the brain of your voice agent. It understands what callers say and decides how to respond. Bolna lets you switch between models like GPT-5.4 mini, Claude and DeepSeek by changing the provider and model in your agent LLM settings, so you can tune for speed, cost or reasoning depth.

Browse

Text-to-Speech

Text-to-speech turns your agent responses into spoken audio. Voice quality shapes how callers perceive your brand. Flat, robotic speech kills trust while natural, expressive voices build it. Bolna integrates with the fastest TTS providers so responses sound human and arrive without awkward pauses.

Browse

Tools & Workflows

Tools let your voice agents take action during a call, not just talk. Book a calendar slot, look up an order through your own API, push a lead into your CRM, or trigger a multi-step automation in Zapier. These integrations turn voice agents from answering machines into workflow engines.

See where ElevenLabs Scribe (STT) fits in your production workflow

Use the demo to walk through provider selection, stack tradeoffs, and the exact workflow you want Bolna to automate.