Speechmatics
for LuumenAI
Check transcription jobs, pull transcripts, and queue new audio for processing
Connect Speechmatics and Luumen can answer questions about your transcription backlog without you opening a browser. List recent jobs in the connected project, check why one failed, pull a finished transcript as JSON, plain text, or SRT, and review Batch usage over a date range. Queuing new audio is a previewed step you approve.
The Speechmatics toolbox
7 tools: 5 read, 2 write. Reads answer instantly. Writes require approval by default. Everything is logged.
- ReadDiscover Speech FeaturesReturn Speechmatics' current Batch languages, locales, domains, translation support, and language-identification capabilities.
- ReadGet TranscriptRetrieve a transcription job's completed transcript as structured JSON v2, plain text, or SRT text.
- ReadGet Transcription JobGet a transcription job's status, configuration, duration, and processing errors, optionally waiting briefly for a terminal state.
- ReadGet UsageGet Batch transcription counts and audio duration aggregated over an optional UTC date range.
- ReadList Transcription JobsList transcription jobs in the connected US1 project, newest first, with an agent-controlled continuation cursor.
- WriteCreate Transcription JobCreate an asynchronous Speechmatics transcription job from a publicly fetchable media URL and return its job ID. Approval by default
- WriteDelete Transcription JobDelete a transcription job and its associated transcript and resources, optionally forcing termination of a running job. Approval by default
One prompt, start to finish
What a governed Speechmatics run looks like inside Luumen.
Questions
Authorize once with API token. Luumen lists the scopes each action needs before you approve the connection, and credentials never appear in the chat.
Read actions answer immediately. Anything that writes — create transcription job, delete transcription job — is shown as a plan and requires approval by default, including the 1 action classified as destructive. Administrators configure that per tool, so you decide exactly which actions can ever run unattended.
You decide. Actions are granted per agent, skill, and team, and per environment — production is not staging. Read access can be broad while writes stay narrow.
Every call to Speechmatics — read or write, approved or declined — is recorded with the actor, the input, and the result, and can be linked to the ticket or change record.
Often connected alongside
All integrations
Adobe Marketing Agent MCPAdobe Experience Platform context, sandbox switches, and async task status
AffindaRead parsed documents, collections, and annotations — change them with approval
Agent MailGive the agent its own mailbox for reading threads and sending updates
AI/ML APIInspect assistants, threads, and runs — and change them as approved stepsPut Speechmatics to work with Luumen
Connect in minutes. Every action scoped, approved, and audited from day one.