Amazon Transcribe
AWS managed speech-to-text with batch + streaming, custom vocabulary, and medical/call-analytics variants.
AWS managed speech-to-text with batch + streaming, custom vocabulary, and medical/call-analytics variants.
Best for teams already on AWS who want IAM-native STT and bundled features like channel ID and PII redaction. Pricing: from $0.0240/min (standard batch, tiered).
What it is
Amazon Transcribe is AWS's managed automatic speech recognition service. It offers batch transcription from S3, real-time streaming over WebSocket and HTTP/2, speaker diarization, channel identification, custom vocabulary, custom language models, and automatic content redaction. Dedicated variants exist for Medical (HIPAA-eligible) and Call Analytics (post-call + real-time analytics with sentiment and issue detection). Pricing is per-second with volume tiers; HIPAA eligibility requires a Business Associate Addendum. Models support 100+ language variants. Available in most AWS commercial regions; some features such as Call Analytics are restricted to a subset of regions.
Watch out for: Per-minute pricing rises with each add-on (PII, custom language model, content redaction); region-specific feature availability.
Install / use
aws transcribe start-transcription-job ...Features
| Speaker diarization | Yes |
| Word-level timestamps | Yes |
| Streaming / real-time | Yes |
| Languages supported | 100 |
| HIPAA eligible | Yes |
Amazon Transcribe vs Whipscribe
| Feature | Amazon Transcribe | Whipscribe |
|---|---|---|
| Category | Transcription APIs | Transcription APIs |
| Pricing | from $0.0240/min (standard batch, tiered) | free beta |
| Speaker diarization | Yes | Yes |
| Word timestamps | Yes | Yes |
| Streaming | Yes | No |
| Languages | 100 | 99 |
| Platforms | API | Web, API, MCP |
Alternatives to Amazon Transcribe
Whipscribe is a managed faster-whisper + whisperX service. If you want transcripts without running infrastructure, paste a URL or drop a file in the form below — you'll have a transcript in seconds.