Amazon Transcribe
aws.amazon.com
- Free plan
- Web
- API
What Amazon Transcribe is
Amazon Transcribe is a managed speech recognition service that converts streaming audio and recorded media into text. It transcribes live streams in real time and processes recorded files stored in Amazon S3 in batches. The service supports 113 languages and can identify the dominant or multiple languages without a language code. Transcript options include punctuation, number formatting, word timestamps, speaker and channel identification, and vocabulary filtering. Users can supply custom vocabulary or train language models with domain-specific text. Call Analytics provides sentiment, issue, speech-characteristic, categorization, and call-summary insights; Transcribe Medical supports dictation and conversational medical transcription and is HIPAA eligible. Controls include PII redaction and toxic-content detection. Transcription outputs can use S3 SSE-S3 or customer-specified AWS KMS encryption at rest, while TLS 1.2 encrypts data in transit. Exports are available as JSON, WebVTT, or SRT, and API access is supported. A free tier includes 60 minutes per month for 12 months, with no rollover. Paid usage starts at $0.01; the listed regional example is billed in one-second increments with no minimum use. The AWS CLI does not support streaming transcription.
Who it suits
It suits teams and developers processing live audio or recordings who need transcripts, speaker labels, timestamps, or language-specific customization. Medical and call workflows may use the listed specialized services.
Good to know, and what to watch
- Handles both live streams and S3 batch files.
- Supports 113 languages and automatic language identification.
- Offers timestamps, speaker labels, and channel identification.
- Exports JSON, WebVTT, and SRT.
- Provides PII redaction and toxic-content detection.
- AWS CLI does not support streaming transcription.
- Free tier is limited to 60 minutes monthly for 12 months.
- Unused free-tier monthly minutes do not roll over.
- Rates vary by AWS Region and volume tier.
Plans
Every plan and priceFree tier
Standard Batch
Standard Streaming
Standard pricing
The verdict
Amazon Transcribe covers real-time and batch transcription, with options for customization, analytics, and privacy controls. Check regional rates and the CLI streaming limitation when choosing how to run it.
What the maker publishes
- API access
- Yes source
- AWS service integrations
- The product page identifies Amazon Connect Contact Lens and AWS HealthScribe as services used with Amazon Transcribe for call analytics and clinical documentation. source
- Call analytics
- Amazon Transcribe Call Analytics provides sentiment, issue, speech-characteristic, categorization, and generative call-summary insights. source
- Compliance
- Amazon Transcribe is covered under AWS HIPAA eligibility and BAA requirements. source
- Customization
- Users can add custom vocabulary and train custom language models with domain-specific text. source
- Encryption
- Transcription outputs can use Amazon S3 SSE-S3 or customer-specified AWS KMS keys at rest, while TLS 1.2 encrypts data in transit. source
- Export formats
- JSON,WebVTT,SRT source
- Free plan
- No source
- Input modes
- The service processes live audio streams for real-time transcription and recorded media files stored in Amazon S3 for batch transcription. source
- Integrations
- Batch transcription uses Amazon S3, and AWS provides SDKs and CLI access for supported programming languages. source
- Languages supported
- 113 source
- Language identification
- Amazon Transcribe can automatically identify the dominant language or multiple languages in audio without a specified language code. source
- Languages
- Amazon Transcribe provides features across 100+ languages. source
- Medical use
- Amazon Transcribe Medical supports dictation and conversational medical transcription and is HIPAA eligible. source
- Privacy controls
- Amazon Transcribe can filter vocabulary, redact personally identifiable information, and detect toxic audio content. source
- Purpose
- Amazon Transcribe is a fully managed automatic speech recognition service that converts streaming and recorded speech to text. source
- Speaker identification
- Yes source
- Timestamp support
- Yes source
- Transcript features
- It supports automatic punctuation, number formatting, word timestamps, speaker identification, channel identification, and vocabulary filtering. source
- Usage limit
- The AWS CLI does not support streaming transcriptions. source
Checked against the maker's pages on 1 Oct 2026.