The objective of this work is to develop effective multiview semi-supervised machine learning strategies for sentence boundary classification problem when only small sets of sentence boundary labeled data are available. We propose three-view and committee-based learning strategies incorporating with co-training algorithms with agreement, disagreement, and self-combined learning strategies using prosodic, lexical and morphological information. We compare experimental results of proposed three-view and committee-based learning strategies to other semi-supervised learning strategies in the literature namely, self-training and co-training with agreement, disagreement, and self-combined strategies. The experiment results show that sentence segmentation performance can be highly improved using multi-view learning strategies that we propose since data sets can be represented by three redundantly sufficient and disjoint feature sets. We show that the proposed strategies substantially improve the average performance when only a small set of manually labeled data is available for Turkish and English spoken languages, respectively.

Sentence segmentation of speech aims at determining sentence boundaries in a stream of words as output by the speech recognizer. Typically, statistical methods are used for sentence segmentation. However, they require significant amounts of labeled data, preparation of which is time-consuming, labor-intensive, and expensive. This work investigates the application of multi-view semi-supervised learning algorithms on the sentence boundary classification problem by using lexical and prosodic information. The aim is to find an effective semi-supervised machine learning strategy when only small sets of sentence boundary-labeled data are available. We especially focus on two semi-supervised learning approaches, namely, self-training and co-training. We also compare different example selection strategies for co-training, namely, agreement and disagreement. Furthermore, we propose another method, called self-combined, which is a combination of self-training and co-training. The experimental results obtained on the ICSI Meeting (MRDA) Corpus show that both multi-view methods outperform self-training, and the best results are obtained using co-training alone. This study shows that sentence segmentation is very appropriate for multi-view learning since the data sets can be represented by two disjoint and redundantly sufficient feature sets, namely, using lexical and prosodic information. Performance of the lexical and prosodic models is improved by 26% and 11% relative, respectively, when only a small set of manually labeled examples is used. When both information sources are combined, the semi-supervised learning methods improve the baseline F-Measure of 69.8% to 74.2%.

This paper presents novel methods for generative, discriminative, and hybrid sequence classification for segmentation of Turkish word sequences into sentences. In the literature, this task is generally solved using statistical models that take advantage of lexical information among others. However, Turkish has a productive morphology that generates a very large vocabulary, making the task much harder. In this paper, we introduce a new set of morphological features, extracted from words and their morphological analyses. We also extend the established method of hidden event language modeling (HELM) to factored hidden event language modeling (fHELM) to handle morphological information. In order to capture non-lexical information, we extract a set of prosodic features, which are mainly motivated from our previous work for other languages. We then employ discriminative classification techniques, boosting and conditional random fields (CRFs), combined with fHELM, for the task of Turkish sentence segmentation.

We investigate the application of the co-training learning algorithm on the sentence boundary classification problem by using lexical and prosodic information. Co-training is a semi-supervised machine learning algorithm that uses multiple weak classifiers with a relatively small amount of labeled data and incrementally uses unlabeled data. The assumption in co-training is that the classifiers can co-train each other, as one can label samples that are difficult for the other. The sentence segmentation problem is very appropriate for the co-training method since it satisfies the main requirements of the co-training algorithm: the dataset can be described by two disjoint and natural views that are redundantly sufficient. In our case, the feature sets are capturing lexical and prosodic information. The experimental results on the ICSI Meeting (MRDA) corpus show the effectiveness of the co-training algorithm for this task. Index Terms: co-training, sentence segmentation, prosody, self-training, Boosting

In this work, a new mathematical modeling approach is proposed for the representation of the speech and audio signals. This approach is based on the generation of the so called predefined signature sequence (PSS) and predefined envelope sequence (PES) sets. After the generation process of the PSS and PES sets, they are clustered by effective k-means clustering algorithm and the PSS and PES are redefined by using these centroids. By using this approach, the drawbacks by means of the size of the sets, speed of the reconstruction process (computational complexity) which arise in the proposed methods previously are highly eliminated. In spite of these improvements, the initial results proved that, the quality of the reconstructed signals remains within the limitations of the acceptable hearing quality.

In this paper, the method of speech modeling which is called SYMPES is introduced and it is compared with the commercially available methods. It is shown that for the same compression ratio or better, SYMPES yields considerably better hearing quality over the coders such as G.726 at 16 Kbps and voice excited LPC-10E of 2.4Kbps.

In this paper, we analyze the effect of model adaptation for dialog act tagging. The goal of adaptation is to improve the performance of the tagger using out-of-domain data or models. Dialog act tagging aims to provide a basis for further discourse analysis and understanding in conversational speech. In this study we used the ICSI meeting corpus with high-level meeting recognition dialog act (MRDA) tags, that is, question, statement, backchannel, disruptions, and floor grabbers/holders. We performed controlled adaptation experiments using the Switchboard (SWBD) corpus with SWBD-DAMSL tags as the out-of-domain corpus. Our results indicate that we can achieve significantly better dialog act tagging by automatically selecting a subset of the Switchboard corpus and combining the confidences obtained by both in-domain and out-of-domain models via logistic regression, especially when the in-domain data is limited.