Share information by removing language barriers with Amazon Translate
summary
Demonstrates using Amazon Transcribe + Amazon Translate to turn Open WHO videos into searchable subtitles, cutting subtitle production for a 30‑minute video from one person a day per language to days to minutes by automating batch transcription, HTML-aware translation and reinserting timestamps. Adds custom vocabulary, custom terminology and Active Custom Translation so the system learns post‑edits, reduces word error rate, and ships an open source Media Insights Engine framework for replication.
AI-written summary · about
chapters (10)
- 0:00Introduction to talk
- 1:11Language barriers in COVID-19
- 5:33Speed vs accuracy trade-off
- 7:56Automating subtitles with AI
- 10:28Accuracy challenges in automation
- 12:20Customizing AI for better results
- 15:03Demo of subtitle workflow
- 23:32WHO’s perspective on translation
- 29:18Amazon Translate capabilities
- 30:26Closing and next steps
related talks
33mMaximize lifetime value of media content
45mTranscend Language Barriers with Custom Machine Learning APIs: um...
21mBoosting Inclusivity: Closed Captioning & Translations in WebRTC
11mInterview
43mBreaking the language barrier: how do we quickly add multilanguage support in our AI application?
20mMeta's AI Translations Inference
22mGemini: Slashing Subtitle Time & Cost with AI
01h 34mReinventing Speech-to-Text Transcriptions with Go and Whisper