Hello, I have been working on designs for two different projects, where both of them would need to use the IBM Watson streaming ASR service.
Based on our discussion at AstriDevCon, I know there is currently no support for that. However, there may be some workarounds I am not aware of.
Would it be possible to write out the audio frames as they get recorded?
Watson supports 16 bit signed little endian audio, IIRC, but there are a few other raw formats available.
The goal is to eventually fund the development of a small module to get something a bit better working, but for now I would be happy with a solution I can build a prototype on.
Thanks to everybody!