[Integrations](https://spiral.ai/integrations) › [Codex](https://spiral.ai/integrations/codex) › [ElevenLabs Scribe v2](https://spiral.ai/integrations/codex/elevenlabs-scribe-v2)

# How to Use ElevenLabs Scribe v2 in Codex

ElevenLabs Scribe v2 turns recorded interviews, meetings, and lectures into text with word timestamps and speaker labels. Its multilingual transcription and vocabulary hints suit recordings with technical terms or product names.

|  |  |
| --- | --- |
| Model | ElevenLabs Scribe v2 |
| Available through | fal |
| Category | Audio |
| Price | $0.0104 per minute of input audio, each started minute counted ($0.01352 with key terms). |
| Returns | json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with ElevenLabs Scribe v2 in Codex

- Use ElevenLabs Scribe v2 to transcribe recordings/customer-interview.mp3 with speaker labels. Use the product names in README.md as vocabulary hints, then create research/interview.md with timestamped quotes and a separate list of product requests.
- Transcribe the 8-minute audio recording in media/tutorial.wav with ElevenLabs Scribe v2. Use the returned word timestamps to create a draft SRT file, and keep laughter tags when they occur.
- Use ElevenLabs Scribe v2 on recordings/design-review.m4a with speaker labels. Write a Markdown decision log with a timestamp for each decision, and leave ownership unassigned when the recording does not establish it.
- Transcribe recordings/api-lecture.mp3 using ElevenLabs Scribe v2, with OpenTelemetry, PostgreSQL, and Kubernetes as vocabulary hints. Save the transcript and create timestamped study notes in docs/training.md.
- Use ElevenLabs Scribe v2 to transcribe the 12-minute bilingual interview in research/interview.wav. Preserve the spoken languages, then add a transcript viewer to the existing app that seeks the audio when a word is clicked.

## How to set up ElevenLabs Scribe v2 in Codex

Pick the way that suits you. Each takes about two minutes.

### Use Spiral's MCP server

1. **Add the hosted server**

```bash
codex mcp add spiral --url https://api.spiral.ai/mcp
```

2. **Sign in**: Codex opens the browser: sign in to Spiral and click Authorize. No API key. `codex mcp list` then shows Spiral.

```bash
codex mcp login spiral
```

### Use the Spiral skill

1. **Send Codex this message**: Your agent reads Spiral's skill and sets itself up.

```
set up https://spiral.ai/SKILL.md
```

2. **Sign in or paste a key**: Approve Spiral in the browser, or give the agent a key from the console.

### Use the Spiral CLI

1. **Install and sign in**: Installs the Spiral CLI, signs you in and adds its skill to your coding agents.

```bash
npx -y spiral-cli@latest init
```

### Call the Spiral API

1. **Create an API key**: In the console, under API keys.

2. **Run ElevenLabs Scribe v2**: One request; the result comes back by polling or a webhook.

```bash
curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "fal",
    "model": "fal-ai/elevenlabs/speech-to-text/scribe-v2",
    "input": {
      "audio_url": "https://storage.googleapis.com/falserverless/example_inputs/elevenlabs/scribe_v2_in.mp3"
    }
  }'
```

Then paste one of the prompts above: Codex finds ElevenLabs Scribe v2 on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/codex

## ElevenLabs Scribe v2 tools and prices

- [ElevenLabs Speech to Text - Scribe V2](https://spiral.ai/tools/fal/fal-ai/elevenlabs/speech-to-text/scribe-v2) (fal): $0.0104 per minute of input audio, each started minute counted ($0.01352 with key terms).

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| language_code |  | — | Language code of the audio |
| tag_audio_events |  | true | Tag audio events like laughter, applause, etc. |
| diarize |  | true | Whether to annotate who is speaking |
| keyterms |  | [] | Words or sentences to bias the model towards transcribing. Up to 100 keyterms, max 50 characters each. Adds 30% premium over base transcription price. |

## What can you do with ElevenLabs Scribe v2 in Codex?

Turn project recordings into searchable transcripts, caption drafts, or timestamped research notes. Your coding agent can use Scribe v2's word timings and speaker labels to build those features around the project's files.

### Interview quote finding

Transcribe recorded interviews with speaker labels, then have your agent list useful quotes with timestamps for checking against the audio.

### Meeting decision records

Turn meeting audio into a speaker-labeled transcript. Have your agent extract decisions and action items with timestamps supporting each entry.

### Podcast caption drafts

Use word timestamps to build draft captions from podcast audio. Include laughter and applause tags where they help convey the recording.

### Technical lecture notes

Supply specialized terms before transcribing lecture audio, then ask your agent to organize the transcript into study notes with timestamp references.

### Multilingual research interviews

Transcribe interviews that switch languages in one recording. Preserve the original-language transcript before asking your agent for a separate translation or summary.

### Recorded support call review

Separate speakers in recorded calls, then have your agent collect customer questions and product mentions with timestamps. Supply product spellings as vocabulary hints.

### Searchable voice notes

Convert spoken notes into searchable text. Have your agent organize the transcript by topic while preserving timestamps back to the original recording.

## Overview

- [fal](https://spiral.ai/providers/fal)
- [All Codex guides](https://spiral.ai/integrations/codex)

## ElevenLabs Scribe v2 in other agents

- [Amp](https://spiral.ai/integrations/amp/elevenlabs-scribe-v2)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/elevenlabs-scribe-v2)
- [Claude](https://spiral.ai/integrations/claude-desktop/elevenlabs-scribe-v2)
- [Claude Code](https://spiral.ai/integrations/claude-code/elevenlabs-scribe-v2)
- [Cline](https://spiral.ai/integrations/cline/elevenlabs-scribe-v2)
- [Cursor](https://spiral.ai/integrations/cursor/elevenlabs-scribe-v2)
- [Devin](https://spiral.ai/integrations/devin/elevenlabs-scribe-v2)
- [dots](https://spiral.ai/integrations/dots/elevenlabs-scribe-v2)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/elevenlabs-scribe-v2)
- [Gemini](https://spiral.ai/integrations/gemini/elevenlabs-scribe-v2)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/elevenlabs-scribe-v2)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/elevenlabs-scribe-v2)
- [goose](https://spiral.ai/integrations/goose/elevenlabs-scribe-v2)
- [Grok](https://spiral.ai/integrations/grok/elevenlabs-scribe-v2)
- [Grok Build](https://spiral.ai/integrations/grok-build/elevenlabs-scribe-v2)
- [Hermes](https://spiral.ai/integrations/hermes/elevenlabs-scribe-v2)
- [Instinct](https://spiral.ai/integrations/instinct/elevenlabs-scribe-v2)
- [Manus](https://spiral.ai/integrations/manus/elevenlabs-scribe-v2)
- [Muse](https://spiral.ai/integrations/muse/elevenlabs-scribe-v2)
- [Muse Code](https://spiral.ai/integrations/muse-code/elevenlabs-scribe-v2)
- [n8n](https://spiral.ai/integrations/n8n/elevenlabs-scribe-v2)
- [Notion](https://spiral.ai/integrations/notion/elevenlabs-scribe-v2)
- [OpenClaw](https://spiral.ai/integrations/openclaw/elevenlabs-scribe-v2)
- [opencode](https://spiral.ai/integrations/opencode/elevenlabs-scribe-v2)
- [Perplexity](https://spiral.ai/integrations/perplexity/elevenlabs-scribe-v2)
- [Raycast](https://spiral.ai/integrations/raycast/elevenlabs-scribe-v2)
- [VS Code](https://spiral.ai/integrations/vscode/elevenlabs-scribe-v2)
- [Windsurf](https://spiral.ai/integrations/windsurf/elevenlabs-scribe-v2)
- [Zapier](https://spiral.ai/integrations/zapier/elevenlabs-scribe-v2)

## More Codex guides

- [ElevenLabs Scribe v1](https://spiral.ai/integrations/codex/elevenlabs-scribe-v1)
- [ElevenLabs Sound Effects](https://spiral.ai/integrations/codex/elevenlabs-sound-effects)
- [ElevenLabs Music v2](https://spiral.ai/integrations/codex/elevenlabs-music-v2)
- [ElevenLabs Sound Effects V2](https://spiral.ai/integrations/codex/elevenlabs-sound-effects-v2)
- [ElevenLabs Music](https://spiral.ai/integrations/codex/elevenlabs-music)
- [ElevenLabs TTS Turbo v2.5](https://spiral.ai/integrations/codex/elevenlabs-tts-turbo-v2-5)

## Questions about ElevenLabs Scribe v2 in Codex

**Can Codex use ElevenLabs Scribe v2?**

Yes. Add Spiral to Codex as an MCP server, and Codex can find and run ElevenLabs Scribe v2 like any built-in tool.

**How much does ElevenLabs Scribe v2 cost in Codex?**

$0.0104 per minute of input audio, each started minute counted ($0.01352 with key terms). You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Codex plan is separate.

**Do I need a fal account to use ElevenLabs Scribe v2 in Codex?**

No. Spiral runs ElevenLabs Scribe v2 on its own fal account. You sign in to Spiral once and never paste a key.

**How long does it take to set up ElevenLabs Scribe v2 in Codex?**

About two minutes: add the hosted server, then sign in.

**Does Codex wait while ElevenLabs Scribe v2 runs?**

ElevenLabs Scribe v2 runs in the background. Codex starts the run, checks on it and gives you the json when it's ready.

**What is Scribe v2 useful for?**

Use it to transcribe recorded interviews, meetings, lectures, and podcasts. Word timestamps help you find passages in the audio, while speaker labels separate contributions in conversations.

**Does Scribe v2 translate recordings into English?**

No. It transcribes speech in the language spoken and can handle multiple languages within one recording. Ask your agent to translate the resulting transcript afterward if needed.

**Can Scribe v2 identify people by name?**

Speaker diarization separates voices using anonymous labels and supports up to 32 speakers. Keep those labels until you can verify which person each label represents.

**How can I improve transcription of unusual names or technical terms?**

Supply the expected spellings as keyterms, such as product names, surnames, and technical phrases. Scribe v2 uses the surrounding audio to decide whether those terms apply, so treat them as vocabulary hints.

**Can Scribe v2 include laughter and applause?**

Yes. Keep audio event tagging enabled when those sounds matter to captions or interview interpretation.

