Rev AI
Affordable, accurate, easy-to-use speech-to-text solutions powered by people and A.I working together. Rev offers transcripts, captions, subtitles, and more.
Verdict
Common use cases
- Transcribe recorded meetings for searchable notes
- Generate captions for video content at scale
- Extract speaker-labeled dialogue from interviews
- Build custom vocabularies for medical or legal terms
- Archive and delete old transcription jobs in bulk
Integration
- Vendor
- Rev AI
- Category
- other
- Auth
- API_KEY
- Tools
- 11
- Composio slug
rev
Tools
- Delete Custom Vocabularydestructive
Tool to delete a completed custom vocabulary and its data. Use when you need to remove an unused vocabulary after confirming it's no longer needed.
- Delete Job By IDdestructive
Tool to delete a completed transcription job and its data. Use when you need to permanently remove a finished job after confirming it's no longer needed.
- Get Account
Tool to retrieve developer account details. Use after authenticating with Rev AI.
- Get Captions
Tool to retrieve captions (SRT or VTT) for a completed Rev.ai transcription job. Use after confirming the job status is 'completed'.
- Get Custom Vocabulary Details
Tool to retrieve custom vocabulary processing details. Use when needing to fetch the status and submitted phrases for a specific custom vocabulary after creation.
- Get Job by ID
Tool to fetch details of a transcription job by its ID. Use when confirming job status and metadata are accurate.
- Get List of Jobs
Tool to get list of transcription jobs from the past 30 days. Use when you need to retrieve and paginate through recent transcription tasks.
- Get Transcript by ID
Tool to retrieve the transcript of a completed Rev.ai job. Use after confirming job is complete. Supports JSON format (with timestamps and speaker info) or plain text format.
- Start Stream Transcription
Tool to start a WebSocket transcription stream. Use when you need real-time speech-to-text streaming via Rev.ai.
- Submit Custom Vocabulary
Tool to submit a custom vocabulary for improved speech recognition. Use when you want to process domain-specific terms asynchronously.
- Submit Transcription Job
Tool to submit a new transcription job. Use when you have a media URL or file bytes ready for async processing.
Setup
Setup guide
- 11. In Switchy, open Settings → Integrations and find Rev AI under 'Other'. 2. Click Connect and paste your Rev AI API key (get it from the Rev AI developer dashboard at rev.ai/access_token). 3. Switchy validates the key by calling Get Account; if it succeeds you'll see 'Connected' with your account email. 4. Open any Space and type '@Rev AI get list of jobs' to confirm the connection works — you should see recent transcription jobs or an empty list if you haven't run any yet. 5. To transcribe a file, upload it to the Space (or reference a public URL), then @mention Rev AI with a prompt like '@Rev AI submit this audio for transcription'. The MCP returns a job ID you can poll for status. 6. Once status is 'completed', retrieve the transcript with '@Rev AI get transcript by ID <job_id>' or captions with '@Rev AI get captions <job_id> format srt'. 7. If you work with specialized terminology, create a custom vocabulary first ('@Rev AI create custom vocabulary with phrases <list>') and reference it when submitting jobs.
What teammates see: by default, memories from Rev AI are scoped to the Space (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Spaces have used this MCP together with a given model.
How Switchy teams use it
Starter prompts
Transcribe Meeting Recording
@Rev AI submit the attached audio file for transcription, then check the job status every 30 seconds until it's done and show me the final transcript in plain text format.Open in a Space →
Generate SRT Captions
@Rev AI get captions for job ID <job_id> in SRT format so I can add subtitles to the video.Open in a Space →
List Recent Transcription Jobs
@Rev AI show me all transcription jobs from the past 30 days, including their status and duration.Open in a Space →
Create Industry Vocabulary
@Rev AI create a custom vocabulary named 'MedicalTerms' with these phrases: angioplasty, myocardial infarction, echocardiogram, and confirm it's ready.Open in a Space →
Clean Up Old Jobs
@Rev AI list all completed jobs older than 20 days, then delete the ones I confirm are no longer needed.Open in a Space →
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@revai list all transcription jobs from the last week and show me which ones are still processing
Found 8 transcription jobs from the past 7 days. 6 are completed: 'team-standup-jan15.mp3' (3m 42s), 'client-call-acme.mp4' (28m 11s), 'product-demo-v2.mov' (12m 5s), 'interview-candidate-3.wav' (45m 22s), 'webinar-recording.mp4' (1h 3m 18s), and 'sales-pitch-draft.m4a' (8m 33s). 2 are still in progress: 'board-meeting-jan16.mp4' (submitted 12 minutes ago, estimated completion in 8-10 minutes) and 'podcast-episode-47.wav' (submitted 3 minutes ago, estimated completion in 15-18 minutes). All completed jobs are ready for transcript retrieval.
This example shows Rev AI's job listing capability, which returns jobs from the past 30 days only. The MCP filters and presents status information clearly. Users should know that job metadata persists for 30 days after creation, but transcripts can be deleted earlier if needed. Pagination is automatic for accounts with high job volumes.
@revai get the full transcript for job ID Umx5c6F7pH7r including speaker labels and timestamps, then summarize the key decisions made
Retrieved transcript for 'team-standup-jan15.mp3' (3m 42s, 2 speakers). Speaker 1 opened discussing sprint progress, noting the API integration is 80% complete. Speaker 2 raised a blocker with the payment gateway at timestamp 1:24, suggesting a vendor switch. At 2:15, Speaker 1 agreed to schedule a technical review meeting for Friday. At 2:58, both confirmed the sprint demo would proceed as planned on Thursday afternoon, excluding the payment feature. Key decisions: (1) Technical review meeting Friday to evaluate payment gateway alternatives, (2) Sprint demo Thursday without payment integration, (3) API integration remains on track for sprint completion.
This example demonstrates transcript retrieval with structured metadata (speakers, timestamps) and pairs it with AI reasoning to extract decisions. The job ID must reference a completed transcription—jobs still processing will return an error. Rev AI's speaker diarization quality depends on audio clarity and distinct voices. Timestamps enable precise reference to conversation moments.
@revai delete the transcription job for 'old-webinar-2023.mp4' and confirm it's been removed from my account
Successfully deleted transcription job ID Bkn8d2M9xR4p ('old-webinar-2023.mp4'). The job and all associated data—including the transcript, captions, and metadata—have been permanently removed from your Rev AI account. This job will no longer appear in your job list or count toward your storage. Note: This action cannot be undone. If you need this transcript in the future, you'll need to re-submit the audio file for transcription.
This example shows the delete operation, which is permanent and irreversible. Users should confirm they have exported any needed transcripts before deletion. The MCP can only delete completed jobs—jobs still processing must finish or be cancelled first. Deleting jobs helps manage account storage and removes sensitive transcription data when no longer needed.
Use-case deep-dives
When Rev AI beats manual note-taking for recurring meetings
A 12-person startup records its Friday all-hands on Zoom and needs searchable transcripts for remote teammates across time zones. Rev AI's Get Transcript by ID tool pulls speaker-labeled JSON with timestamps, so the team can quote exact moments in Notion or Linear without re-watching 40 minutes of video. The Get List of Jobs tool surfaces the last 30 days, making it easy to link past decisions in Slack threads. This works if your team actually reviews transcripts—if people just skim summaries, you're paying for accuracy you won't use. The Delete Job By ID tool keeps storage tidy after 90 days. If your all-hands runs under 20 people and you reference past calls more than twice a month, Rev AI is the right call.
Rev AI for user research when you need verbatim quotes
A 4-person product team conducts 8 customer interviews per sprint and needs to pull exact quotes into Figma or Miro boards during synthesis sessions. Rev AI's transcript format includes timestamps, so designers can jump to the moment a user said 'I never click that button' without scrubbing through recordings. The Custom Vocabulary tools let you pre-load product-specific terms so 'onboarding flow' doesn't transcribe as 'on boarding flaw'. This setup wins if your research process depends on verbatim evidence—if you're just tagging themes, a cheaper summarization tool is enough. The 30-day job retention means you need to export key quotes to your research repo before they expire. If you run discovery sprints and quote customers in roadmap decks, Rev AI justifies the API cost.
When Rev AI makes sense for compliance-light call review
A 6-person support team records 40 calls per week and the lead spot-checks 5 for coaching feedback. Rev AI's Get Captions tool outputs SRT files that sync with the video player, so the manager can scan the transcript and jump to moments where the rep missed a step. The Get Account tool confirms usage limits before the month ends. This works if you're doing lightweight QA, not regulated compliance—Rev AI doesn't flag PII or sentiment, so you're still reading manually. If your call volume exceeds 200 per month, the per-minute cost adds up fast and you should consider a purpose-built QA platform. For small teams doing occasional coaching reviews without compliance overhead, Rev AI keeps it simple and you only pay for what you transcribe.
Frequently asked
What does the Rev AI MCP do in Switchy?
It connects your Rev AI account so Switchy can submit audio files for transcription, check job status, retrieve completed transcripts with timestamps and speaker labels, and manage custom vocabularies. You can automate transcription workflows without leaving Switchy or writing API code yourself.
Do I need a Rev AI API key to use this MCP?
Yes. You'll need to generate an API key from your Rev AI developer account and paste it into Switchy's connection flow. The key authenticates all requests, so anyone on your team who connects it needs their own key or you share one account-wide key with appropriate access controls.
Can the Rev AI MCP transcribe live audio or only uploaded files?
Only uploaded files. The MCP submits pre-recorded audio to Rev AI's transcription service and retrieves results once processing finishes. If you need real-time transcription, you'll have to use Rev AI's streaming API directly—this MCP doesn't expose that.
How is this different from uploading files to Rev AI's website?
The MCP lets you script transcription inside Switchy workflows—submit a batch of files, check statuses, pull transcripts into documents, all without context-switching. The website is manual one-at-a-time. Use the MCP when you're automating; use the website when you're doing a single ad-hoc job.
Who on my team should connect the Rev AI MCP?
Whoever manages your Rev AI account or runs transcription workflows. They'll need the API key and enough Rev AI credits to cover usage. Transcription jobs consume your Rev AI balance, not Switchy's—so coordinate with whoever owns the billing relationship with Rev AI.