Figure 5
A sample waveform (clean) and its pitches detected by SPTK in various SNR situations. The threshold of voiced-unvoiced classification was set to 6.0 (SPTK default). For the frames detected as unvoiced, SPTK outputs zero. The test data was prepared by blending noise at different SNRs. The noise was recorded in a car moving on an expressway with a fan at a medium level.