Yobe’s software documentation includes an introduction to the Yobe SDK with library requirements, input audio data requirements and sampling rates. There are four SDK Variants which generate three specific outputs:
ASR Listening: Optimized signals for machine consumption
Identification Listening: Speaker-specific signals ideal for managing human
crosstalk or scenarios where only a specific/known speaker(s) voice input is wanted.
Identification Indexing: Real-time (frame by frame) Speaker Identification