Speech source separation is essential for speech-related applications because this process enhances the input speech signal for the main processing model. Most of the current approaches for this task focus on separating the speech of commonly high-frequency noises or a particular background sound. They cannot clear the signals which intersect with the human speech in its frequency range. To deal with this problem, we propose a hybrid approach combining a variational autoencoder (VAE) and a bandpass filter (BPF). This method can extract and enhance the speech signal in the mixture of many elements such as speech signal, the high-frequency noises, and many kinds of different background sounds which interfere with the speech sound. Experimental results showed that our model can extract effectively the speech signal with 15.02 dB in Signal to Interference Ratio (SIR) and 12.99 dB in Signal to Distortion Ratio (SDR). On the other hand, we can adjust the passband to identify the range of frequency at the output signal to apply for a particular application like gender classification.

Speech separation plays an important role in a speech-related system because it can denoise, extract and enhance speech signal, and after all improve the accuracy and performance of the system. In recent years, many approaches only separate the speech out of commonly high-frequency noise or a particular background sound. We propose a more powerful approach, combining an autoencoder and a bandpass filter to separate speech signals. This combination can extract the speech in the mixture with not only high-frequency noise but also many kinds of different background sounds. Our approach can be flexibly applied for the new background sounds. Experimental results show that our model can extract fastly and effectively the speech signal with 9.01 dB in SIR and 11.26 in SDR. On the other hand, we can adjust the passband to identify the range of frequency at the output signal to apply for particular applications.

No abstract

No abstract