I have a set of fairly robust workflows I've developed using Node Red to interface with various parts of the Watson toolset. I'd like to create a voice-via-phone interface to them via Twilio, but I'm having a hard time figuring out exactly what I need to do to access the live audio feed.
Via Node-Red, what Watson STT requires:
The Speech To Text converts the human voice into the written word. This service uses machine intelligence to combine information about grammar and language structure with knowledge of the composition of the audio signal to generate a more accurate transcription
The audio file to be analysed should be passed in on msg.payload.
Supported msg.payload types:.
String URL to audio Buffer Raw Audio Bytes Audio must be a WAV, FLAC or OGG encoded file.
The question is this: I've scoured the API for Programmable Voice, but haven't seemed to be successful in locating what API call I would initiate to create an audio stream (ref'd via URL or Audio Bytes) that can be piped into the Watson STT.
Anyone who's more experienced than I with the Twilio toolset with some answers?
