No abstract

Speaker extraction aims to extract the target speech signal from a multi-talker environment given a target speaker's reference speech. We recently proposed a time-domain solution, SpEx, that avoids the phase estimation in frequency-domain approaches. Unfortunately, SpEx is not fully a time-domain solution since it performs time-domain speech encoding for speaker extraction, while taking frequency-domain speaker embedding as the reference. The size of the analysis window for timedomain and the size for frequency-domain input are also different. Such mismatch has an adverse effect on the system performance. To eliminate such mismatch, we propose a complete time-domain speaker extraction solution, that is called SpEx+. Specifically, we tie the weights of two identical speech encoder networks, one for the encoder-extractor-decoder pipeline, another as part of the speaker encoder. Experiments show that the SpEx+ achieves 0.8dB and 2.1dB SDR improvement over the state-of-the-art SpEx baseline, under different and same gender conditions on WSJ0-2mix-extr database respectively.

Most existing AU detection works considering AU relationships are relying on probabilistic graphical models with manually extracted features. This paper proposes an end-to-end deep learning framework for facial AU detection with graph convolutional network (GCN) for AU relation modeling, which has not been explored before. In particular, AU related regions are extracted firstly, latent representations full of AU information are learned through an auto-encoder. Moreover, each latent representation vector is feed into GCN as a node, the connection mode of GCN is determined based on the relationships of AUs. Finally, the assembled features updated through GCN are concatenated for AU detection. Extensive experiments on BP4D and DISFA benchmarks demonstrate that our framework significantly outperforms the state-of-the-art methods for facial AU detection. The proposed framework is also validated through a series of ablation studies.

A blind dereverberation method based on power spectral subtraction (SS) using a multi-channel least mean squares algorithm was previously proposed to suppress the reverberant speech without additive noise. The results of isolated word speech recognition experiments showed that this method achieved significant improvements over conventional cepstral mean normalization (CMN) in a reverberant environment. In this paper, we propose a blind dereverberation method based on generalized spectral subtraction (GSS), which has been shown to be effective for noise reduction, instead of power SS. Furthermore, we extend the missing feature theory (MFT), which was initially proposed to enhance the robustness of additive noise, to dereverberation. A one-stage dereverberation and denoising method based on GSS is presented to simultaneously suppress both the additive noise and nonstationary multiplicative noise (reverberation). The proposed dereverberation method based on GSS with MFT is evaluated on a large vocabulary continuous speech recognition task. When the additive noise was absent, the dereverberation method based on GSS with MFT using only 2 microphones achieves a relative word error reduction rate of 11.4 and 32.6% compared to the dereverberation method based on power SS and the conventional CMN, respectively. For the reverberant and noisy speech, the dereverberation and denoising method based on GSS achieves a relative word error reduction rate of 12.8% compared to the conventional CMN with GSS-based additive noise reduction method. We also analyze the effective factors of the compensation parameter estimation for the dereverberation method based on SS, such as the number of channels (the number of microphones), the length of reverberation to be suppressed, and the length of the utterance used for parameter estimation. The experimental results showed that the SS-based method is robust in a variety of reverberant environments for both isolated and continuous speech recognition and under various parameter estimation conditions.

Previous studies of speech emotion recognition using either empirical features (e.g., F0, energy, and voice probability) or spectrogram-based statistical features. The empirical features can highlight the human knowledge of emotion recognition, while the statistical features enable a general representation, but they do not emphasize human knowledge sufficiently. However, the use of these two kinds of features together can complement some features that may be unconsciously used by humans in daily life but have not been realized yet. Based on this consideration, this paper proposes a dynamic fusion framework to utilize the potential advantages of the complementary spectrogram-based statistical features and the auditory-based empirical features. In addition, a kernel extreme learning machine (KELM) is adopted as the classifier to distinguish emotions. To validate the proposed framework, we conduct experiments on two public emotional databases, including Emo-DB and IEMOCAP databases. The experimental results demonstrate that the proposed fusion framework significantly outperforms the existing state-of-the-art methods. The results also show that the proposed method, by integrating the auditory-based features with spectrogram-based features, could achieve a notably improved performance over the conventional methods. INDEX TERMS Speech emotion recognition, auditory-based features, spectrogram-based features, complementary features, kernel extreme learning machine.