Presign upload
Request a private S3 upload URL for an MP3 or M4A file up to 200 MB.
POST /v1/uploads/presign
Built from the original Super Transcriber workflow
Super Transcriber API
A privacy-conscious transcription API for teams that need async audio jobs, private AWS deployment, usage controls, and a clean REST interface their product can call today.
- Async
- REST jobs
- Private
- S3 uploads
- Optional
- Customer AWS
From app to platform
The original transcription experience, repackaged for your product.
Super Transcriber started as an end-user-facing web app for uploading audio and receiving transcripts. This API-first edition turns that workflow into a ready-to-integrate backend for developers, internal platforms, and companies that want transcription inside their own existing systems without building the AWS pipeline from scratch.
API workflow
Four calls from audio file to transcript.
The hosted API keeps integration simple, while the enterprise path deploys the same Terraform stack into a customer-owned AWS account.
Start job
Submit the returned S3 key and optional speaker or duration metadata.
POST /v1/transcriptions
Track status
Poll by job id, list recent jobs with cursor pagination, or receive a webhook.
GET /v1/transcriptions/{jobId}
Retrieve transcript
Download TXT, normalized JSON, SRT, or VTT with private, customer-scoped access.
COMPLETED
Developer experience
A straightforward integration surface for product teams.
Every endpoint uses structured JSON, API-key authentication, durable async jobs, and clear error bodies. Customers can start on the hosted API and move to private AWS deployment when procurement or data controls require it.
- API keys are hashed before storage and scoped to customer-owned jobs.
- Usage limits are enforced before Transcribe work starts.
- Stripe Checkout and webhooks are available without being required.
curl -sS -X POST "$API_BASE_URL/v1/transcriptions" \
-H "x-api-key: $SUPER_TRANSCRIBER_API_KEY" \
-H "content-type: application/json" \
-d '{
"s3Key": "customers/cust_123/uploads/job_123/meeting.mp3",
"fileName": "meeting.mp3",
"durationSeconds": 600,
"speakerCount": 2
}'import { SuperTranscriberClient } from "@super-transcriber/api-client";
const client = new SuperTranscriberClient({
baseUrl: process.env.SUPER_TRANSCRIBER_API_URL!,
apiKey: process.env.SUPER_TRANSCRIBER_API_KEY!
});
const job = await client.createTranscription({
s3Key: presign.s3Key,
fileName: "meeting.mp3",
durationSeconds: 600,
speakerCount: 2
});{
"event": "transcription.completed",
"jobId": "job_97b15fe3",
"status": "COMPLETED",
"customerId": "cust_123",
"transcriptText": "Welcome everyone..."
}AWS-native platform
Built for low idle cost and enterprise deployment options.
Serverless core
Amazon Transcribe handles speech recognition and speaker diarization. API Gateway HTTP API, Lambda Node.js 24 on arm64, SQS, EventBridge, and DynamoDB coordinate the workflow.
Private media path
Audio enters through presigned S3 uploads. Buckets block public access and lifecycle rules expire files and generated transcripts.
Usage-aware billing
Per-customer active job and monthly minute limits are enforced before transcription starts. Stripe is optional and disabled until configured.
Security controls
Hashed API keys, scoped job ownership checks, per-Lambda IAM roles, HMAC webhooks, structured errors, and no raw API-key logging.
Enterprise path
The same Terraform stack can be deployed into a customer AWS account for teams that need direct infrastructure ownership.
Cost-first shape
No VPC, NAT Gateway, RDS, OpenSearch, or fixed monthly compute. Idle costs include minimal secrets, storage, queue polling, and scheduled recovery. Enrichment polling stops when enrichment is off.
Packaging
Monetization-ready without forcing a fixed platform bill.
Plans are represented as API usage limits and can be wired to Stripe Checkout, private contracts, or customer-owned deployments.
Starter API
For prototypes, internal tools, and first integrations.
- 120 monthly transcription minutes
- 5 active jobs
- 200 MB file limit
- API-key access and polling
Team API
For products adding transcription to customer workflows.
- Higher monthly minute pools
- Webhook delivery
- Optional transcript enrichment worker
- Stripe-ready billing hooks
This TypeScript worker can call an LLM through Amazon Bedrock after transcription. It is disabled on the hosted deployment.
Private AWS
For companies that want the same system in their account.
- Terraform deployment package
- Customer-owned S3 and DynamoDB
- Enterprise security review path
- No shared media storage
Ready-to-integrate backend