I really believe these science advancements will define next generation audio and video communication capabilities. Super excited about adding powerful machine learning and artificial intelligence capabilities to real-time communication workloads. Many thanks to the excellent support we got from the Transcribe team led by Vasi, Vikram, Pavan, Priyank, Katrin, and Stefano. ![]() Finally, we take the complexity out of the developer experience - enabling builders to automatically leverage Transcribe and Chime’s global footprint.Įxcellent work John, Arvindh, Prem, Noah, Hoang, Haifeng, Michael, Umut, and the many others who helped launch this super useful feature for our customers. Because we work with individual audio streams, we can capture information that may get lost to the human ear because of cross talk or varying speaker volume. With just a few API calls you can add low latency transcription to your app - powering capabilities ranging from accessibility to sentiment and topic analysis with Comprehend. ![]() What’s neat is we use an advanced signal processing algorithm to detect the active talkers on a multi-party audio session, stream these individual audio sessions before it is mixed to Transcribe or Transcribe Medical, and then take the resulting low-latency, high accuracy captions and surface them to an app developer. The integration of the Amazon Chime SDK with Transcribe is what powers captions for Slack Huddles. The team and I are super excited to launch this feature in partnership with the Transcribe team.
0 Comments
Leave a Reply. |