The Predictability of Phonemes in Signals

The figure shows time on the x-axis and then the explainability of various features like nasal, vowel, voicing, approximate, fricative. And so that's basically seven phonemes per second. In these stories, each phoneme is about 80 milliseconds or so. So therefore, if a phoneme is being instantiated in the signal for half a second, that means that you're going to have three phonemes at a time. Right? The prediction performance is not high. It's only slightly above chance.