Singel Channel Source Separation
The here presented studies were carried out within the basic project Robust, part of COAST. Robust deals with robust speech pre-processing for speech recognition. The development of a new method for source separation is the task within the speech enhancement module of Robust.
As a first step, I carried out Single Channel Source Separation using synthetic a priori known excitation signals and searched for the vocal tract envelope (VTE) information. Some examples can be found here:VTE-based Singel Channel Source Separation
- First results for the source-filter based single channel source speartion can be found here:Source-Filter Based Single Sensor Speech Separation.
- Source-filter based single channel speech separation using multi-pitch tracking examples.