STT is now able to mark hesitation in the transcription. Not seen anywhere in the documentation whether this is controlled by a parameter setting. If it is, then the STT node needs to be able to set the parameter. Either way the node should provide a mechanism to filter out the hesitation in the output, if not needed.
STT is now able to mark hesitation in the transcription. Not seen anywhere in the documentation whether this is controlled by a parameter setting. If it is, then the STT node needs to be able to set the parameter. Either way the node should provide a mechanism to filter out the hesitation in the output, if not needed.
https://www.ibm.com/watson/developercloud/doc/speech-to-text/output.shtml#hesitation