I need to implement an app, where we do speech to text of audio files uploaded by the user.Requirement:- User uploads an audio file and specifies which transcription engine to use (google speech to text, nuance)- User receives the speech to text response.- Assume I can have other REST api's as well.How to:Is it good to have a separate multipart upload api to upload audio files that returns a file token and another transcript api which receives file token and engine type as params. Or do the file upload in the same transcript request?How to implement the app so that other engines can be added easily?How can we make this handle 1000's of requests?How to build microservices on this, so that the high processing modules can easily have multiple instances?Thanks a lot in advance.
Submitted July 27, 2019 at 11:01AM by stopcharla
No comments:
Post a Comment