Source-filter model in current research

[Chad M]

How is the source-filter model incorporated into modern speech synthesis technologies like HMM and DNN systems? Do these systems choose from a library of individual, pre-generated acoustic sounds and then arrange the sounds to improve things like intonation, stress, and rhythm?

[Simon]

This is a little beyond the Speech Processing course, but is covered fully in the more advanced Speech Synthesis course.

The short answer is that most statistical parametric speech synthesisers (whether HMM or DNN) use a source-filter model to generate the waveform. The HMM or DNN predicts the parameters of the source (e.g., F0) and of the filter (e.g., its frequency response).

[Author]

Posts