[Integrations](https://spiral.ai/integrations) › [opencode](https://spiral.ai/integrations/opencode) › [Insanely Fast Whisper with video](https://spiral.ai/integrations/opencode/insanely-fast-whisper-with-video)

# How to Use Insanely Fast Whisper with video in opencode

Insanely Fast Whisper with video turns recorded speech into transcripts or English translations using Whisper Large v3. Its video support extracts the audio, making it useful for interviews, lectures, and caption drafts.

|  |  |
| --- | --- |
| Model | Insanely Fast Whisper with video |
| Available through | Replicate |
| Category | Audio |
| Price | $0.00127 per second of processing. |
| Returns | json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with Insanely Fast Whisper with video in opencode

- Use Insanely Fast Whisper with video to transcribe the English narration in docs/demo.wav with chunk timestamps. Compare the transcript with our README and propose corrections supported by the narration.
- Use Insanely Fast Whisper with video to transcribe media/tutorial.mp3 with word timestamps. Build an SRT draft in the project's captions directory and flag its timing for review.
- Use Insanely Fast Whisper with video to transcribe recordings/design-review.wav in its original language. Then draft a Markdown note separating explicit decisions from suggestions, with timestamp references.
- Use Insanely Fast Whisper with video to translate the Spanish speech in recordings/walkthrough.mp3 into English. Then draft a documentation outline from the narration and list screen details that still need manual review.
- Use Insanely Fast Whisper with video to transcribe recordings/bug-report.wav with chunk timestamps. Then compare the reported behavior with the project's code and propose a reproduction checklist.

## How to set up Insanely Fast Whisper with video in opencode

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

### Use Spiral's MCP server

1. **Add the server**: Leave out `--global` to add Spiral to the current project only. No header and no client ID or secret: opencode registers itself with Spiral.

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

2. **Sign in**: Your browser opens. Sign in to Spiral, pick the organization and click Authorize, then return to the terminal. No API key. Inside opencode, `/mcps` starts the same sign-in.

```bash
opencode mcp auth spiral
```

3. **Check it**: Spiral shows as connected.

```bash
opencode mcp list
```

### Use the Spiral skill

1. **Send opencode 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 Insanely Fast Whisper with video**: 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": "replicate",
    "model": "turian/insanely-fast-whisper-with-video",
    "input": {
      "audio": "<audio>"
    }
  }'
```

Then paste one of the prompts above: opencode finds Insanely Fast Whisper with video on Spiral, shows the price and runs it.

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

## Insanely Fast Whisper with video tools and prices

- [Insanely Fast Whisper (with video)](https://spiral.ai/tools/replicate/turian/insanely-fast-whisper-with-video) (Replicate): $0.00127 per second of processing.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| url |  | — | Video URL for yt-dlp to download the audio from. Either this or audio must be provided. |
| task | transcribe, translate | transcribe | Task to perform: transcribe or translate to another language. (default: transcribe). |
| audio |  | — | Audio file. Either this or url must be provided. |
| language |  | — | Optional. Language spoken in the audio, specify None to perform language detection. |
| timestamp | chunk, word | chunk | Whisper supports both chunked as well as word level timestamps. (default: chunk). |
| batch_size |  | 64 | Number of parallel batches you want to compute. Reduce if you face OOMs. (default: 64). |

## What can you do with Insanely Fast Whisper with video in opencode?

Turn recorded explanations into documentation and caption drafts within your project. Your coding agent can use the transcript to propose edits and keep timestamp references for review.

### Podcast transcript drafts

Transcribe an episode, then have your agent organize the text into readable sections. Check passages selected for quotation against the audio.

### Caption preparation

Request word timestamps for a recorded tutorial, then have your agent assemble a caption draft for timing review.

### Interview quote research

Transcribe a recorded interview and use timestamps to locate candidate quotations. Verify the selected wording before using it.

### English notes from recorded talks

Translate non-English speech into English text, then ask your agent to prepare notes with references back to the recording.

### Searchable lecture archives

Transcribe recorded lectures in their spoken language, then index the transcripts with timestamp references for finding relevant passages.

### Recorded product demo documentation

Transcribe a demo's narration, then have your agent draft documentation from the spoken explanation. Review screen-only details separately.

### Voice memo organization

Turn spoken notes into transcripts, then ask your agent to sort them into ideas, questions, and proposed tasks.

## Overview

- [Replicate](https://spiral.ai/providers/replicate)
- [All opencode guides](https://spiral.ai/integrations/opencode)

## Insanely Fast Whisper with video in other agents

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

## More opencode guides

- [Index TTS 2.0](https://spiral.ai/integrations/opencode/index-tts-2-0)
- [Inworld TTS-1.5 Max](https://spiral.ai/integrations/opencode/inworld-tts-1-5-max)
- [HeyGen Voice Turbo](https://spiral.ai/integrations/opencode/heygen-voice-turbo)
- [Kling Create Voice](https://spiral.ai/integrations/opencode/kling-create-voice)
- [HeyGen Voice](https://spiral.ai/integrations/opencode/heygen-voice)
- [Kling TTS](https://spiral.ai/integrations/opencode/kling-tts)

## Questions about Insanely Fast Whisper with video in opencode

**Can opencode use Insanely Fast Whisper with video?**

Yes. Add Spiral to opencode as an MCP server, and opencode can find and run Insanely Fast Whisper with video like any built-in tool.

**How much does Insanely Fast Whisper with video cost in opencode?**

$0.00127 per second of processing. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your opencode plan is separate.

**Do I need a Replicate account to use Insanely Fast Whisper with video in opencode?**

No. Spiral runs Insanely Fast Whisper with video on its own Replicate account. You sign in to Spiral once and never paste a key.

**How long does it take to set up Insanely Fast Whisper with video in opencode?**

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

**Does opencode wait while Insanely Fast Whisper with video runs?**

Insanely Fast Whisper with video runs in the background. opencode starts the run, checks on it and gives you the json when it's ready.

**Can Insanely Fast Whisper translate speech into any language?**

Speech translation produces English text. Use transcription when you want to keep the recording's original language, then ask your agent to translate that transcript separately if needed.

**Does Insanely Fast Whisper understand what happens on screen?**

Video transcription uses the extracted audio. It cannot tell you what a silent demonstration shows or read text that appears only on screen.

**How accurate is Insanely Fast Whisper across languages?**

Accuracy varies by language, and Whisper documents weaker results for languages with less training data. Review a sample in the language you need before processing a large archive.

**When should I use word timestamps?**

Choose word timestamps when you need to find individual words for caption editing or clip selection. Chunk timestamps suit a reading transcript with broader reference points.

**Can I publish an Insanely Fast Whisper transcript without checking it?**

Review it first: Whisper can insert words that were never spoken. Check quoted passages against the recording before publication.

