Speech marks annotate the audio data with metadata about the synthesis process, like word timing or phoneme details.
Speech marks annotate the audio data with metadata about the synthesis process, like word timing or phoneme details.