AWS ai documentation change
Summary
Removed references to Amazon Transcribe RAI Performance Assessment Report availability
Security assessment
The change only removes documentation about requesting a performance report. No security vulnerabilities, weaknesses, or security feature updates are mentioned or implied.
Diff
diff --git a/ai/responsible-ai/transcribe-speech-recognition/overview.md b/ai/responsible-ai/transcribe-speech-recognition/overview.md index c0c4f8d39..e0ef0ffbc 100644 --- a//ai/responsible-ai/transcribe-speech-recognition/overview.md +++ b//ai/responsible-ai/transcribe-speech-recognition/overview.md @@ -224 +224 @@ As with all AWS AI services, numerous stakeholders are involved in Amazon Transc -Customers who are required to validate AI or ML models as part of their governance processes may find the following information relevant. Amazon Transcribe performs speech-to-text conversion and does not generate novel content. The model assumes that input audio contains naturally occurring human speech in a supported language, recorded at a supported sample rate and format. Customers can use custom vocabularies, PII redaction, vocabulary filtering, confidence score thresholds, and human oversight as compensating controls to address model limitations for their specific use case. Word-level confidence scores are recalibrated during each model update cycle to ensure that scores accurately reflect transcription correctness and that confidence score distributions remain consistent, supporting the stability of downstream workflows that depend on specific thresholds. Amazon Transcribe is monitored by automated canary systems that periodically evaluate accuracy, latency, and other metrics against reference datasets, with alerts triggered on significant drift and rollback mechanisms in place. (See Model Updates and Ongoing Monitoring for details.) Customers should implement their own monitoring appropriate to their use case, including periodic retesting on representative data, tracking of output quality metrics, and review of edge cases. (See Performance Drift under Deployment and Performance Optimization Best Practices.) Customers requiring detailed technical information beyond this Service Card – including evaluation dataset descriptions, quantitative performance results with confidence intervals, and risk assessment results – may request the Amazon Transcribe RAI Performance Assessment Report through their AWS account team, subject to a non-disclosure agreement. +Customers who are required to validate AI or ML models as part of their governance processes may find the following information relevant. Amazon Transcribe performs speech-to-text conversion and does not generate novel content. The model assumes that input audio contains naturally occurring human speech in a supported language, recorded at a supported sample rate and format. Customers can use custom vocabularies, PII redaction, vocabulary filtering, confidence score thresholds, and human oversight as compensating controls to address model limitations for their specific use case. Word-level confidence scores are recalibrated during each model update cycle to ensure that scores accurately reflect transcription correctness and that confidence score distributions remain consistent, supporting the stability of downstream workflows that depend on specific thresholds. Amazon Transcribe is monitored by automated canary systems that periodically evaluate accuracy, latency, and other metrics against reference datasets, with alerts triggered on significant drift and rollback mechanisms in place. (See Model Updates and Ongoing Monitoring for details.) Customers should implement their own monitoring appropriate to their use case, including periodic retesting on representative data, tracking of output quality metrics, and review of edge cases. (See Performance Drift under Deployment and Performance Optimization Best Practices.) @@ -294,2 +293,0 @@ As a managed AI service, Amazon Transcribe models are continuously improved by A - * For detailed performance data, evaluation dataset descriptions, and risk assessment results, customers may request the Amazon Transcribe RAI Performance Assessment Report through their AWS account team, subject to a non-disclosure agreement. -