Speechmatics for AI agents
Speechmatics turns audio and video into text through asynchronous transcription jobs, used by teams that need accurate transcripts at scale rather than live captions. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.
What agents do in Speechmatics.
Submit a recording for transcription
Create a transcription job for a recorded call or video, returning a text transcript once it finishes.
Check a job's status before fetching it
Get a transcription job's status to confirm it's complete before pulling the finished transcript.
Confirm before deleting a transcription job
Review a job's transcript before deleting it, since removing the job also removes the stored transcript.
7 Speechmatics actions, graded by risk.
Every Speechmatics action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
4Look things up. Allowed by default.
- speechmatics.get_usageGet Usage
- speechmatics.get_transcriptGet Transcript
- speechmatics.get_transcription_jobGet Transcription Job
- speechmatics.list_transcription_jobsList Transcription Jobs
write
2Create and change things. Allow, or ask the user first.
- speechmatics.create_transcription_jobCreate Transcription Job
- speechmatics.discover_speech_featuresDiscover Speech Features
destructive
1Delete, cancel or archive. Ask first, or deny outright.
- speechmatics.delete_transcription_jobDelete Transcription Job
Speechmatics in three steps.
- 01Your users connect SpeechmaticsThey add their Speechmatics api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, writes like “create Transcription Job” can wait for the user, and “delete Transcription Job” can be denied outright.
- 03Any agent can actYour agent calls Speechmatics through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('speechmatics', { read: 'allow', write: 'ask', // create_transcription_job destructive: 'deny', // delete_transcription_job }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Speechmatics connects.
Users add their Speechmatics api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
The same Speechmatics connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →
- AUTH
- API key
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use Speechmatics from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraph
Speechmatics and Arc0, answered.
Can I use Speechmatics with Claude, ChatGPT or Cursor?
Yes. Connect Speechmatics to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the Speechmatics actions you allow.
How do users connect Speechmatics?
Users add their Speechmatics api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
Which Speechmatics actions can my agent take?
7 in total: 4 read, 2 write and 1 destructive, such as “create Transcription Job”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in Speechmatics?
Yes. Actions like “delete Transcription Job” are graded destructive. Set destructive actions to deny, or to ask so the user approves each one, and blocked calls still show up on the audit log.
Can my own backend call Speechmatics too?
Yes. The same Speechmatics connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Speechmatics into your agent.
Your users connect Speechmatics once, under your brand. Your agent gets 7 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan