Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Digital processing of speech signals

Results of the ICDAR 2015 Robust Reading Competition are presented. A new Challenge 4 on Incidental Scene Text has been added to the Challenges on Born-Digital Images, Focused Scene Images and Video Text. Challenge 4 is run on a newly acquired dataset of 1,670 images evaluating Text Localisation, Word Recognition and End-to-End pipelines. In addition, the dataset for Challenge 3 on Video Text has been substantially updated with more video sequences and more accurate ground truth data. Finally, tasks assessing End-to-End system performance have been introduced to all Challenges. The competition took place in the first quarter of 2015, and received a total of 44 submissions. Only the tasks newly introduced in 2015 are reported on. The datasets, the ground truth specification and the evaluation protocols are presented together with the results and a brief summary of the participating methods.

/pdf/icdar-2015-competition-on-robust-reading-17zul2skj0.pdf

ICDAR 2015 competition on Robust Reading

This report presents the final results of the ICDAR 2013 Robust Reading Competition. The competition is structured in three Challenges addressing text extraction in different application domains, namely born-digital images, real scene images and real-scene videos. The Challenges are organised around specific tasks covering text localisation, text segmentation and word recognition. The competition took place in the first quarter of 2013, and received a total of 42 submissions over the different tasks offered. This report describes the datasets and ground truth specification, details the performance evaluation protocols used and presents the final results along with a brief summary of the participating methods.

/pdf/icdar-2013-robust-reading-competition-3sb9m2ga5i.pdf

ICDAR 2013 Robust Reading Competition

Algorithms for extracting epochs or glottal closure instants (GCIs) from voiced speech typically fall into two categories: i) ones which operate on linear prediction residual (LPR) and ii) those which operate directly on the speech signal. While the former class of algorithms (such as YAGA and DPI) tend to be more accurate, the latter ones (such as ZFR and SEDREAMS) tend to be more noise-robust. In this letter, a temporal measure termed the cumulative impulse strength is proposed for locating the impulses in a quasi-periodic impulse-sequence embedded in noise. Subsequently, it is applied for detecting the GCIs from the inverted integrated LPR using a recursive algorithm. Experiments on two large corpora of speech with simultaneous electroglottographic recordings demonstrate that the proposed method is more robust to additive noise than the state-of-the-art algorithms, despite operating on the LPR.

Cumulative Impulse Strength for Epoch Extraction

This paper addresses the problem of Hindi compound word splitting and its relevance to developing a good quality phonetizer for Hindi Speech Synthesis. The constituents of a Hindi compound word are not separated by space or hyphen. Hence, most of the existing compound splitting algorithms can not be applied to Hindi. We propose a new technique for automatic extraction of compound words from Hindi corpus. Preliminary tests conducted on the algorithm have shown a split rate of 92 to 96% of the input compound words. Of these splits, around 83 to 87% are correct splits. A few modifications have been suggested, which will improve the accuracy of the splits. Finally, we observe an improvement of 1.6% in Hindi Grapheme-to-Phoneme (G2P) conversion as a result of using a phonetized compound word lexicon, created by the above technique.

/pdf/automatic-generation-of-compound-word-lexicon-for-hindi-4rxogn2ua4.pdf

Automatic Generation of Compound Word Lexicon for Hindi Speech Synthesis

This paper proposes an automatic acoustic-phonetic method for estimating voice-onset time of stops. This method requires neither transcription of the utterance nor training of a classifier. It makes use of the plosion index for the automatic detection of burst onsets of stops. Having detected the burst onset, the onset of the voicing following the burst is detected using the epochal information and a temporal measure named the maximum weighted inner product. For validation, several experiments are carried out on the entire TIMIT database and two of the CMU Arctic corpora. The performance of the proposed method compares well with three state-of-the-art techniques.

Estimation of voice-onset time in continuous speech using temporal measures

This article describes a postprocessing strategy for online, handwritten, isolated Tamil words. Contributions have been made with regard to two issues hardly addressed in the online Indic word recognition literature, namely, use of (1) language models exploiting the idiosyncrasies of Indic scripts and (2) expert classifiers for the disambiguation of confused symbols.The input word is first segmented into its individual symbols, which are recognized using a primary support vector machine (SVM) classifier. Thereafter, we enhance the recognition accuracy by utilizing (i) a bigram language model at the symbol or character level and (ii) expert classifiers for reevaluating and disambiguating the different sets of confused symbols. The symbol-level bigram model is used in a traditional Viterbi framework. The concept of a character comprising multiple symbols is unique to Dravidian languages such as Tamil. This multi-symbol feature of Tamil characters has been exploited in proposing a novel, prefix-tree-based character-level bigram model that does not use Viterbi search; rather it reduces the search space for each input symbol based on its left context.For disambiguating confused symbols, a dynamic time-warping approach is proposed to automatically identify the parts of the online trace that discriminates between the confused classes. Fine classification of these regions by dedicated expert SVMs reduces the extent of confusions between such symbols. The integration of segmentation, prefix-tree-based language model and disambiguation of confused symbols is presented on a set of 15,000 handwritten isolated online Tamil words. Our results show recognition accuracies of 93.0% and 81.6% at the symbol and word level, respectively, as compared to the baseline classifier performance of 88.4% and 65.1%, respectively.

Bigram Language Models and Reevaluation Strategy for Improved Recognition of Online Handwritten Tamil Words

N-gram language models and lexicon-based word-recognition are popular methods in the literature to improve recognition accuracies of online and offline handwritten data. However, there are very few works that deal with application of these techniques on online Tamil handwritten data. In this paper, we explore methods of developing symbol-level language models and a lexicon from a large Tamil text corpus and their application to improving symbol and word recognition accuracies. On a test database of around 2000 words, we find that bigram language models improve symbol (3%) and word recognition (8%) accuracies and while lexicon methods offer much greater improvements (30%) in terms of word recognition, there is a large dependency on choosing the right lexicon. For comparison to lexicon and language model based methods, we have also explored re-evaluation techniques which involve the use of expert classifiers to improve symbol and word recognition accuracies.

A. G. Ramakrishnan

Papers

Cumulative Impulse Strength for Epoch Extraction

Automatic Generation of Compound Word Lexicon for Hindi Speech Synthesis

Estimation of voice-onset time in continuous speech using temporal measures

Bigram Language Models and Reevaluation Strategy for Improved Recognition of Online Handwritten Tamil Words

Language models for online handwritten Tamil word recognition