Abstract-Broadcast News (BN) transcription has been a challenging research area for many years. In the last couple of years the availability of large amounts of roughly transcribed acoustic training data and advanced model training techniques has offered the opportunity to greatly reduce the error rate on this task. This paper describes the design and performance of BN transcription systems which make use of these developments. First the effects of using lightly-supervised training data and advanced acoustic modelling techniques are discussed. The design of a real-time broadcast news recognition system is then detailed using these new models. As system combination has been found to yield large gains in performance, a range of frameworks that allow multiple recognition outputs to be combined are next described. These include the use of multiple types of acoustic models and multiple segmentations. As a contrast a system developed by multiple sites allowing cross-site combination, the "SuperEARS" system, is also described. The various models and recognition configurations are evaluated using several recent BN development and evaluation test sets. These new BN transcription systems can give gains of over 25% relative to the CU-HTK 2003 BN system.

Electrocardiogram (ECG) signals are usually corrupted by baseline wander, power-line interference, muscle noise etc. Numerous methods have been proposed to remove these noises. However, in case of wireless recording of the ECG signal it gets corrupted by the additive white Gaussian noise (AWGN). For the correct diagnosis, removal of AWGN from ECG signals becomes necessary as it affects the diagnostic features. The natural signals exhibit correlation among their samples and this property has been exploited in various signal restoration tasks. Motivated by that, in this study we propose a non-local wavelet transform domain ECG signal denoising method which exploits the correlations among both local and non-local samples of the signal. In the proposed method, the similar blocks of the samples are grouped in a matrix and then denoising is achieved by the shrinkage of its two-dimensional discrete wavelet transform coefficients. The experiments performed on a number of ECG signals show significant quantitative and qualitative improvements in denoising performance over the existing ECG signal denoising methods.

No abstract

This paper discusses the development of the CU-HTK Mandarin Broadcast News (BN) transcription system. The Mandarin BN task includes a significant amount of English data. Hence techniques have been investigated to allow the same system to handle both Mandarin and English by augmenting the Mandarin training sets with English acoustic and language model training data. A range of acoustic models were built including models based on Gaussianised features, speaker adaptive training and feature-space MPE. A multi-branch system architecture is described in which multiple acoustic model types, alternate phone sets and segmentations can be used in a system combination framework to generate the final output. The final system shows state-of-the-art performance over a range of test sets.

In the presented work, we explore some of the challenges in recognizing children's speech on automatic speech recognition (ASR) systems developed using adults' speech. In such mismatched ASR tasks, a severely degraded recognition performance is observed due to the gross mismatch in the acoustic attributes between those two groups of speakers. Among the various sources of mismatch, we focus on the large differences in the average pitch values across the adult and child speakers in this work. Earlier studies have shown that the Mel-filterbank employed in the feature extraction is not able to smooth out the pitch harmonics sufficiently in particularly for the high-pitched child speakers. As a result of that, the acoustic features derived for the adult and the child speakers turn out to be significantly mismatched. For addressing this problem, we propose a simple technique based on adaptive-liftering for deriving the pitch-robust features. This enables us to reduce the sensitivity of the acoustic features to the gross variations in pitch across the speakers. The proposed features are found to result in improved performance in the context of deep neural network based ASR system. Further with the use of the existing feature normalization techniques, additional gains are noted.