Two discriminative methods for solving tone problems in Mandarin speech recognition are presented. First, discriminative training on the HMM （hidden Markov model） based tone models is proposed. Then an integration t...Two discriminative methods for solving tone problems in Mandarin speech recognition are presented. First, discriminative training on the HMM （hidden Markov model） based tone models is proposed. Then an integration technique of tone models into a large vocabulary continuous speech recognition system is presented. Discriminative model weight training based on minimum phone error criteria is adopted aiming at optimal integration of the tone models. The extended Baum Welch algorithm is applied to find the model-dependent weights to scale the acoustic scores and tone scores. Experimental results show that tone recognition rates and continuous speech recognition accuracy can be improved by the discriminatively trained tone model. Performance of a large vocabulary continuous Mandarin speech recognition system can be further enhanced by the discriminatively trained weight combinations due to a better interpolation of the given models.展开更多
Valeo, involved in engine cooling fan system design for many years, is interested in noise prediction tools for axial fans. Thus, this paper describes a two-part study of tonal noise computation. The first part deals ...Valeo, involved in engine cooling fan system design for many years, is interested in noise prediction tools for axial fans. Thus, this paper describes a two-part study of tonal noise computation. The first part deals with the prediction of tonal noise using analytical models. As for the second part, it describes a hybrid approach for predicting tonal noise where the sources are extracted from an Unsteady Reynolds-Averaged Naviers-Stocks （URANS） simulation and then propagated into the far, free field using the Ffowcs Williams and Hawkings＇ acoustic analogy. The computational domain is meshed with 46 million polyhedral elements and the simulation takes into account the exact geometry of the rotor blades, the stator blades and the shroud. The results from the first part show that analytical models can be used for comparisons between different fan geometries, but are unable to provide accurate noise predictions compared to experimental results. The simulation shows non-periodic blade loading over a whole fan revolution, and different blade loading between the blades. This introduces some bias in the assessment of the acoustic performance of the fan. Overall, the results from the hybrid method are in accordance with the experimental results.展开更多
This letter proposes an effective and robust speech feature extraction method based on statistical analysis of Pitch Frequency Distributions (PFD) for speaker identification. Compared with the conventional cepstrum, P...This letter proposes an effective and robust speech feature extraction method based on statistical analysis of Pitch Frequency Distributions (PFD) for speaker identification. Compared with the conventional cepstrum, PFD is relatively insensitive to Additive White Gaussian Noise (AWGN), but it does not show good performance for speaker identification, even if under clean environments. To compensate this shortcoming, PFD and conventional cepstrum are combined to make the ultimate decision, instead of simply taking one kind of features into account.Experimental results indicate that the hybrid approach can give outstanding improvement for text-independent speaker identification under noisy environments corrupted by AWGN.展开更多
文摘Two discriminative methods for solving tone problems in Mandarin speech recognition are presented. First, discriminative training on the HMM （hidden Markov model） based tone models is proposed. Then an integration technique of tone models into a large vocabulary continuous speech recognition system is presented. Discriminative model weight training based on minimum phone error criteria is adopted aiming at optimal integration of the tone models. The extended Baum Welch algorithm is applied to find the model-dependent weights to scale the acoustic scores and tone scores. Experimental results show that tone recognition rates and continuous speech recognition accuracy can be improved by the discriminatively trained tone model. Performance of a large vocabulary continuous Mandarin speech recognition system can be further enhanced by the discriminatively trained weight combinations due to a better interpolation of the given models.
文摘Valeo, involved in engine cooling fan system design for many years, is interested in noise prediction tools for axial fans. Thus, this paper describes a two-part study of tonal noise computation. The first part deals with the prediction of tonal noise using analytical models. As for the second part, it describes a hybrid approach for predicting tonal noise where the sources are extracted from an Unsteady Reynolds-Averaged Naviers-Stocks （URANS） simulation and then propagated into the far, free field using the Ffowcs Williams and Hawkings＇ acoustic analogy. The computational domain is meshed with 46 million polyhedral elements and the simulation takes into account the exact geometry of the rotor blades, the stator blades and the shroud. The results from the first part show that analytical models can be used for comparisons between different fan geometries, but are unable to provide accurate noise predictions compared to experimental results. The simulation shows non-periodic blade loading over a whole fan revolution, and different blade loading between the blades. This introduces some bias in the assessment of the acoustic performance of the fan. Overall, the results from the hybrid method are in accordance with the experimental results.
文摘This letter proposes an effective and robust speech feature extraction method based on statistical analysis of Pitch Frequency Distributions (PFD) for speaker identification. Compared with the conventional cepstrum, PFD is relatively insensitive to Additive White Gaussian Noise (AWGN), but it does not show good performance for speaker identification, even if under clean environments. To compensate this shortcoming, PFD and conventional cepstrum are combined to make the ultimate decision, instead of simply taking one kind of features into account.Experimental results indicate that the hybrid approach can give outstanding improvement for text-independent speaker identification under noisy environments corrupted by AWGN.
