How to Use Kotoba STT in opencode
Kotoba STT turns recorded speech into text, with a focus on Japanese and East Asian languages. Use it for clips up to two minutes when you need a transcript, word timings, or speaker labels.
Prompts to try with Kotoba STT in opencode
Copy an example into opencode, or ask in your own words.
Use Kotoba STT to transcribe the 90-second Japanese demo recording in this project with word timings. Create a draft SRT file and preserve the raw transcript beside it.
Use Kotoba STT to transcribe the 60-second English feature walkthrough in this project. Use product names from the README as spelling guidance, then compare the transcript with the documented behavior.
Use Kotoba STT on the 90-second Japanese interview excerpt in this repository with speaker separation and word timings. Save a transcript organized by speaker and time.
Use Kotoba STT to transcribe the 45-second Korean bug-report recording in this project. Draft a reproduction checklist from the transcript and mark details the recording does not establish.
Use Kotoba STT to transcribe the 60-second Japanese narration in this project with personal names written in katakana. Save the transcript for the localization reviewer.
How to set up Kotoba STT in opencode
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- 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.
opencode mcp add spiral --global --url https://api.spiral.ai/mcp
- 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.
opencode mcp auth spiral
- Check it
Spiral shows as connected.
opencode mcp list
Use the Spiral skill
- Send opencode this message
Your agent reads Spiral's skill and sets itself up.
set up https://spiral.ai/SKILL.md
- Sign in or paste a key
Approve Spiral in the browser, or give the agent a key from the console.
Use the Spiral CLI
- Install and sign in
Installs the Spiral CLI, signs you in and adds its skill to your coding agents.
npx -y spiral-cli@latest init
Call the Spiral API
- Create an API key
In the console, under API keys.
- Run Kotoba STT
One request; the result comes back by polling or a webhook.
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": "kotoba-technologies/kotoba-stt/v1/speech-to-text", "input": { "audio_url": "<audio_url>", "language": "ja" } }'
Then paste one of the prompts above: opencode finds Kotoba STT on Spiral, shows the price and runs it.
Kotoba STT providers and prices
Spiral runs Kotoba STT through fal. Open a card for today's price, with every fee included.
Settings you can ask for
Kotoba STT works from your audio URL. Ask opencode for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Language | ja, en, ko, zh, es, zh-hk, zh-tw, or auto | Language spoken in the audio. | |
| With timestamps | on or off | off | Return |
| Diarization | on or off | off | Attribute words to speakers ( |
| Kana | on or off | off | Write personal and place names in katakana (Japanese). |
Every API field of Kotoba STT · Speech to Text is on its tool page: open its card above.
What can you do with Kotoba STT in opencode?
Turn recorded demos, interviews, and voice notes into text your coding agent can use alongside project files. Ask the agent to turn timed transcripts into caption drafts or compare spoken product terminology with the repository.
Japanese interview excerpts
Transcribe an interview excerpt of up to two minutes with speaker labels and timings, then have your agent arrange the transcript for editorial review.
Caption drafts for short videos
Transcribe a video's spoken audio in clips of up to two minutes with word timings. Have your agent turn the timed text into caption drafts for review.
Product demo transcripts
Turn a short narrated demo into text, supplying product names as spelling guidance. Use the transcript as the starting point for written instructions.
Customer research excerpts
Transcribe recorded customer comments in clips of up to two minutes. Keep the transcript beside any themes or observations your agent extracts from it.
Japanese name review
Request katakana personal names in a short Japanese recording to prepare a phonetic transcript for a reviewer.
Recorded voice notes
Transcribe short Japanese, English, Korean, or Chinese voice notes. Have your agent turn the text into a checklist while preserving the original transcript.
More on Kotoba STT and opencode
Overview
Kotoba STT in other agents
Questions about Kotoba STT in opencode
Can opencode use Kotoba STT?
Yes. Add Spiral to opencode as an MCP server, and opencode can find and run Kotoba STT like any built-in tool.
How much does Kotoba STT cost in opencode?
Up to $0.005382 per run at the default inputs for 60 seconds of input. 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 fal account to use Kotoba STT in opencode?
No. Spiral runs Kotoba STT on its own fal account. You sign in to Spiral once and never paste a key.
How long does it take to set up Kotoba STT in opencode?
About two minutes: add the server, then sign in, then check it.
Does opencode wait while Kotoba STT runs?
Kotoba STT runs in the background. opencode starts the run, checks on it and gives you the json when it's ready.
What is Kotoba STT best suited for?
Recorded speech, especially Japanese and East Asian audio. Word timings and speaker labels make it useful for preparing caption drafts and reviewing interview excerpts.
How long can a recording be?
Keep each clip within two minutes. For a longer recording, prepare shorter excerpts before transcription.
Which spoken languages are documented?
Kotoba documents Japanese, English, Korean, and Chinese for file transcription.
How can I guide the spelling of names and technical terms?
Supply the relevant spellings as keywords, such as a speaker's name or a product name. Treat them as spelling guidance and check the resulting transcript against the recording.
Can I get speaker labels with timings?
Yes. Request speaker separation and word timings together when you need to see who spoke and where their words occur in the recording.
Start using Kotoba STT in opencode
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into opencode:
set up https://spiral.ai/SKILL.md


