Amazon Transcribe

• Automatic Speech Recognition (ASR) service
• input = audio; output = text
• language customisation, filters for privacy, audience-appropriate language, speaker identification
• pay as you use (per s)
• Use case:
  • Full text indexing of audio - allow searching
  • meeting notes
  • subtitles/ caption
  • call analytics (characteristics, sentiment, summarization)
  • integration of application