Hello RealSense people,
I've been waiting for this features since I posted the question during the original PercSDK contest over a year ago. So I'm asking again. Does the RealSense Text To Speech API give you phoneme timing maps yet? I'm referring to the timing information you can get with other TTS packages like NeoSpeech, that let you time something like an animated characters mouth movements. That is, a table you get with each Text To Speech generated waveform that tells you, with accuracy, at what millisecond offset a particular phoneme occurs in the generated audio waveform. You can use that information to choose a particular mouth shape for an animated character to "lip sync" the character to the TTS waveform. You didn't have it last year. I really hope you do now! So, is this feature there?
