Lamia Bouafif¹, Noureddine Ellouze²

Sound & Vibration, Vol.52, No.6, pp. 2-10, 2018, DOI:10.32604/sv.2018.02410

Abstract Sound indexing and segmentation of digital documents especially in the internet and digital libraries are very useful to simplify and to accelerate the multimedia document retrieval. We can imagine that we can extract multimedia files not only by keywords but also by speech semantic contents. The main difficulty of this operation is the parameterization and modelling of the sound track and the discrimination of the speech, music and noise segments. In this paper, we will present a Speech/Music/Noise indexing interface designed for audio discrimination in multimedia documents. The program uses a statistical method based on ANN and HMM classifiers. After… More >

Yue Zhao¹, Jianjian Yue¹, Wei Song^1,*, Xiaona Xu¹, Xiali Li¹, Licheng Wu¹, Qiang Ji²

Journal on Internet of Things, Vol.1, No.1, pp. 17-23, 2019, DOI:10.32604/jiot.2019.05866

Abstract We proposed a method using latent regression Bayesian network (LRBN) to extract the shared speech feature for the input of end-to-end speech recognition model. The structure of LRBN is compact and its parameter learning is fast. Compared with Convolutional Neural Network, it has a simpler and understood structure and less parameters to learn. Experimental results show that the advantage of hybrid LRBN/Bidirectional Long Short-Term Memory-Connectionist Temporal Classification architecture for Tibetan multi-dialect speech recognition, and demonstrate the LRBN is helpful to differentiate among multiple language speech sets. More >

Yue Zhao¹, Jianjian Yue¹, Wei Song^1,*, Xiaona Xu¹, Xiali Li¹, Licheng Wu¹, Qiang Ji²

CMC-Computers, Materials & Continua, Vol.60, No.3, pp. 1223-1235, 2019, DOI:10.32604/cmc.2019.05636

Abstract Tibetan language has very limited resource for conventional automatic speech recognition so far. It lacks of enough data, sub-word unit, lexicons and word inventories for some dialects. And speech content recognition and dialect classification have been treated as two independent tasks and modeled respectively in most prior works. But the two tasks are highly correlated. In this paper, we present a multi-task WaveNet model to perform simultaneous Tibetan multi-dialect speech recognition and dialect identification. It avoids processing the pronunciation dictionary and word segmentation for new dialects, while, in the meantime, allows training speech recognition and dialect identification in a single… More >

Zhifeng Wang¹, Diqun Yan^1,*, Rangding Wang¹, Li Xiang¹, Tingting Wu¹

CMC-Computers, Materials & Continua, Vol.56, No.2, pp. 247-259, 2018, DOI: 10.3970/cmc.2018.02902

Abstract Speech resampling is a typical tempering behavior, which is often integrated into various speech forgeries, such as splicing, electronic disguising, quality faking and so on. By analyzing the principle of resampling, we found that, compared with natural speech, the inconsistency between the bandwidth of the resampled speech and its sampling ratio will be caused because the interpolation process in resampling is imperfect. Based on our observation, a new resampling detection algorithm based on the inconsistency of band energy is proposed. First, according to the sampling ratio of the suspected speech, a band-pass Butterworth filter is designed to filter out the… More >