What is the difference between AWS Transcribe > Streaming Transcription feature and Kinesis Video Streams(For Audio Input) for live streaming audio

Question

Hi My requirement is I have live audio stream as input, say a call between 2 people, now to convert that audio to text on live and pick certain keywords from that extracted text and insert in Database.

As per architecture in https://github.com/aws-samples/amazon-connect-realtime-transcription Both AWS Kinesis Video Streams service and AWS Transcribe used for live streaming but as per link : https://aws.amazon.com/blogs/machine-learning/amazon-transcribe-now-supports-real-time-transcriptions/ AWS Transcribe supports live transcription then why in that architecture Kinesis used ?

If any one know, please help me in understanding, Hope Amazon connect can ingest live audio to AWS Transcribe for live transcription.

Answer 1

Amazon Kinesis Video Streams is the service that enables streaming voice data from Amazon Connect. Amazon Transcribe can ingest streams from any source for real-time transcription, but the only way to get that real-time data from Amazon Connect is via Kinesis. The launch announcement for real-time streams might help make this more clear:

With the customer voice stream feature, your customer audio is automatically sent to Amazon Kinesis Video Streams, where it can be accessed by the integrations that you allow. For example, you could integrate customer voice stream with real-time text transcription and sentiment analysis for immediate feedback on call quality, or use this feature with a 3rd party voice biometric product to authenticate the caller automatically without having to enter a password or confirm personal information.