1-4hit |
M. Shahidur RAHMAN Tetsuya SHIMAMURA
This paper explores the potential of pitch determination from bone conducted (BC) speech. Pitch determination from normal air conducted (AC) speech signal can not attain the expected level of accuracy for every voice and background conditions. In contrast, since BC speech is caused by the vibrations that have traveled through the vocal tract wall, it is robust against ambient conditions. Though an appropriate model of BC speech is not known, it has regular harmonic structure in the lower spectral region. Due to this lowpass nature, pitch determination from BC speech is not usually affected by the dominant first formant. Experiments conducted on simultaneously recorded AC and BC speech show that BC speech is more reliable for pitch estimation than AC speech. With little human work, pitch contour estimated from BC speech can also be used as pitch reference that can serve as an alternate to the pitch contour extracted from laryngograph output which is sometimes inconsistent with simultaneously recorded AC speech.
Chiori HORI Bing ZHAO Stephan VOGEL Alex WAIBEL Hideki KASHIOKA Satoshi NAKAMURA
The performance of speech translation systems combining automatic speech recognition (ASR) and machine translation (MT) systems is degraded by redundant and irrelevant information caused by speaker disfluency and recognition errors. This paper proposes a new approach to translating speech recognition results through speech consolidation, which removes ASR errors and disfluencies and extracts meaningful phrases. A consolidation approach is spun off from speech summarization by word extraction from ASR 1-best. We extended the consolidation approach for confusion network (CN) and tested the performance using TED speech and confirmed the consolidation results preserved more meaningful phrases in comparison with the original ASR results. We applied the consolidation technique to speech translation. To test the performance of consolidation-based speech translation, Chinese broadcast news (BN) speech in RT04 were recognized, consolidated and then translated. The speech translation results via consolidation cannot be directly compared with gold standards in which all words in speech are translated because consolidation-based translations are partial translations. We would like to propose a new evaluation framework for partial translation by comparing them with the most similar set of words extracted from a word network created by merging gradual summarizations of the gold standard translation. The performance of consolidation-based MT results was evaluated using BLEU. We also propose Information Preservation Accuracy (IPAccy) and Meaning Preservation Accuracy (MPAccy) to evaluate consolidation and consolidation-based MT. We confirmed that consolidation contributed to the performance of speech translation.
In this paper, we propose a predictive block-constrained trellis-coded quantization (BC-TCQ) to quantize cepstral coefficients for distributed speech recognition. For prediction of the cepstral coefficients, the first order auto-regressive (AR) predictor is used. To quantize the prediction error signal effectively, we use the BC-TCQ. The quantization is compared to the split vector quantizers used in the ETSI standard, and is shown to lower cepstral distance and bit rates.
Shigeki MATSUDA Takatoshi JITSUHIRO Konstantin MARKOV Satoshi NAKAMURA
In this paper, we describe a parallel decoding-based ASR system developed of ATR that is robust to noise type, SNR and speaking style. It is difficult to recognize speech affected by various factors, especially when an ASR system contains only a single acoustic model. One solution is to employ multiple acoustic models, one model for each different condition. Even though the robustness of each acoustic model is limited, the whole ASR system can handle various conditions appropriately. In our system, there are two recognition sub-systems which use different features such as MFCC and Differential MFCC (DMFCC). Each sub-system has several acoustic models depending on SNR, speaker gender and speaking style, and during recognition each acoustic model is adapted by fast noise adaptation. From each sub-system, one hypothesis is selected based on posterior probability. The final recognition result is obtained by combining the best hypotheses from the two sub-systems. On the AURORA-2J task used widely for the evaluation of noise robustness, our system achieved higher recognition performance than a system which contains only a single model. Also, our system was tested using normal and hyper-articulated speech contaminated by several background noises, and exhibited high robustness to noise and speaking styles.