Acoustics of Natural Prompts

摘要

"A good VUI isn't about fooling callers, making them believe they are talking to a real person…… It's about making them forget, that they are talking to a machine" Speech recognition-based systems utilise a Voice User Interface (VUI). The human-machine interaction or dialog is a process of turn taking, where the system output can be generated by synthetic speech (TTS) or is taken care of by pre-recorded prompts. The TTS engine transfers text into synthetic speech at run time, providing directly available prompts. If the system output is covered by pre-recorded prompts, those prompts need to be written, recorded and post-produced first, before they are available for use.