How to Use Universal-3 Pro in Claude Code
Universal-3 Pro transcribes recordings with instructions for terminology, verbatim speech, and mixed-language conversations. Use it when you need to guide how speech becomes text, including whether to retain hesitations and false starts.
Prompts to try with Universal-3 Pro in Claude Code
Copy an example into Claude Code, or ask in your own words.
Use Universal-3 Pro to transcribe recordings/customer-interview.wav. Build a short terminology prompt from the product names in README.md, then save the transcript alongside the recording.
Transcribe recordings/usability-session.wav with Universal-3 Pro. Explicitly retain um, uh, hesitations, repeated words, and false starts so I can study where the participant struggled.
Use Universal-3 Pro on recordings/bilingual-demo.mp3. Preserve English and Spanish exactly as spoken, including switches within a sentence, and save the transcript in research/.
Transcribe recordings/architecture-review.mp3 with Universal-3 Pro. Supply relevant service names from docs/architecture.md as context, then draft a separate list of decisions supported by the transcript.
Use Universal-3 Pro to transcribe recordings/support-call.wav and attempt to tag beeps and hold music. Save the transcript and mark the sound tags for manual checking.
How to set up Universal-3 Pro in Claude Code
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the hosted server
This adds Spiral to the current project. Add --scope user to use it in every project.
claude mcp add --transport http spiral https://api.spiral.ai/mcp
- Sign in
Start claude, type /mcp, choose spiral and Authenticate. Sign in to Spiral in the browser and click Authorize. No API key.
Use the Spiral skill
- Send Claude Code 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 Universal-3 Pro
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": "assemblyai", "model": "universal-3-pro", "input": { "audio_url": "<audio_url>" } }'
Then paste one of the prompts above: Claude Code finds Universal-3 Pro on Spiral, shows the price and runs it.
Universal-3 Pro providers and prices
Spiral runs Universal-3 Pro through AssemblyAI. Open a card for today's price, with every fee included.
$0.273 per hour of audio, per channel. Features, per hour: speaker labels $0.026, key terms $0.065, translation $0.078, speaker identification $0.026, custom formatting $0.039, action items $0.026, summaries $0.026 ($0.091 at medium effort), entity detection $0.104, sentiment $0.026, key phrases $0.013, topic detection $0.195, content moderation $0.195, profanity filter $0.013, PII redaction $0.104, audio redaction $0.065, medical mode $0.195, custom prompt $0.065.
Settings you can ask for
Universal-3 Pro works from your audio URL and prompt. Ask Claude Code for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Domain | medical-v1 | Enable domain-specific transcription models to improve accuracy for specialized terminology. | |
| Punctuate | on or off | on | Enable Automatic Punctuation, can be true or false |
| Redact pii | on or off | off | Redact PII from the transcribed text using the Redact PII model, can be true or false. |
| Format text | on or off | on | Enable Text Formatting. Universal-3 Pro always formats the text: |
| Temperature | 0 to 1 | Control the amount of randomness injected into the model's response. | |
| Audio end at | a whole number | The point in time, in milliseconds, to stop transcribing in your media file. | |
| Disfluencies | on or off | off | Transcribe Filler Words, like "umm", in your media file; can be true or false. |
| Multichannel | on or off | off | Enable Multichannel transcription, can be true or false. |
| Custom topics | on or off | This parameter does not currently have any functionality attached to it. | |
| Content safety | on or off | off | Enable Content Moderation, can be true or false |
| Iab categories | on or off | off | Enable Topic Detection, can be true or false |
| Redact pii sub | entity_name or hash | The replacement logic for detected PII, can be | |
| Speaker labels | on or off | off | Enable Speaker diarization, can be true or false. Requires |
| Auto highlights | on or off | off | Enable Key Phrases, either true or false |
| Audio start from | a whole number | The point in time, in milliseconds, to begin transcribing in your media file. | |
| Entity detection | on or off | off | Enable Entity Detection, can be true or false |
| Filter profanity | on or off | off | Filter profanity from the transcribed text, can be true or false. |
| Redact pii audio | on or off | off | Generate a copy of the original media file with spoken PII "beeped" out, can be true or false. |
| Speech threshold | 0 to 1 | Reject audio files that contain less than this fraction of speech. | |
| Remove audio tags | all or speaker | Universal-3.5 Pro generates rich transcripts that can include inline annotations such as audio event markers and speaker cues. | |
| Speakers expected | a whole number | Tells the speaker label model how many speakers it should attempt to identify. | |
| Language detection | on or off | Automatic language detection identifies the spoken language and routes the request to the best model. | |
| Sentiment analysis | on or off | off | Enable Sentiment Analysis, can be true or false. Requires |
| Redact pii audio quality | mp3 or wav | Controls the filetype of the audio created by redact_pii_audio. | |
| Content safety confidence | 25 to 100 | The confidence threshold for the Content Moderation model. | |
| Redact pii return unredacted | on or off | When set to | |
| Language confidence threshold | 0 to 1 | The confidence threshold for the automatically detected language. |
Every API field of Universal-3 Pro Speech to Text is on its tool page: open its card above.
What can you do with Universal-3 Pro in Claude Code?
Turn recorded interviews and technical discussions into transcripts your project can use. Your coding agent can give Universal-3 Pro terminology from project documentation and use the resulting text to draft research notes.
Technical interview transcripts
Use Universal-3 Pro to transcribe engineering interviews with context about product names and technical terms.
Customer research
Preserve hesitations, repeated words, and false starts with Universal-3 Pro when exact phrasing matters to interview analysis.
Bilingual support calls
Use Universal-3 Pro for English-Spanish calls, asking it to retain each phrase in its spoken language.
Sales coaching
Transcribe recorded sales calls with Universal-3 Pro while retaining fillers and restarts for a coach to review.
Clinical conversation drafts
Give Universal-3 Pro relevant medication names and clinical terminology when preparing a transcript for human review.
Podcast transcript drafts
Use Universal-3 Pro to transcribe episodes with guest names and topic context supplied before transcription.
Call sound review
Test Universal-3 Pro on recorded calls with instructions to tag beeps or hold music. Check its experimental sound tags against the audio.
More on Universal-3 Pro and Claude Code
Overview
Universal-3 Pro in other agents
Questions about Universal-3 Pro in Claude Code
Can Claude Code use Universal-3 Pro?
Yes. Add Spiral to Claude Code as an MCP server, and Claude Code can find and run Universal-3 Pro like any built-in tool.
How much does Universal-3 Pro cost in Claude Code?
$0.273 per hour of audio, per channel. Features, per hour: speaker labels $0.026, key terms $0.065, translation $0.078, speaker identification $0.026, custom formatting $0.039, action items $0.026, summaries $0.026 ($0.091 at medium effort), entity detection $0.104, sentiment $0.026, key phrases $0.013, topic detection $0.195, content moderation $0.195, profanity filter $0.013, PII redaction $0.104, audio redaction $0.065, medical mode $0.195, custom prompt $0.065. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Claude Code plan is separate.
Do I need an AssemblyAI account to use Universal-3 Pro in Claude Code?
No. Spiral runs Universal-3 Pro on its own AssemblyAI account. You sign in to Spiral once and never paste a key.
How long does it take to set up Universal-3 Pro in Claude Code?
About two minutes: add the hosted server, then sign in.
Does Claude Code wait while Universal-3 Pro runs?
Universal-3 Pro runs in the background. Claude Code starts the run, checks on it and gives you the json when it's ready.
Which languages does Universal-3 Pro support?
Universal-3 Pro supports English, Spanish, German, French, Portuguese, and Italian. It can transcribe conversations that mix those languages.
Can Universal-3 Pro keep filler words and false starts?
Yes. Ask explicitly for fillers such as um and uh, hesitations, repeated words, and false starts so the transcript retains the speech patterns you need.
What should I put in a Universal-3 Pro prompt?
Give the recording's topic, relevant names, specialized terminology, and the transcript style you want. Keep the instructions within 1,500 words.
How long can a recording be?
AssemblyAI documents a maximum of 10 hours and 5 GB per batch transcription request, with a minimum duration of 160 milliseconds. Split recordings that exceed either maximum.
Are prompted speaker labels reliable on long recordings?
AssemblyAI warns that prompt-generated labels can become inconsistent when a long recording is processed in chunks. Review speaker assignments across the recording before relying on them.
Start using Universal-3 Pro in Claude Code
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Claude Code:
set up https://spiral.ai/SKILL.md


