Shirin Saleem

Multi-lingual offline handwriting recognition using hidden Markov models: a script-independent approach

This paper introduces a script-independent methodology for multilingual offline handwriting recognition (OHR) based on the use of Hidden Markov Models (HMM). The OHR methodology extends our script-independent approach for OCR of machine-printed text images. The feature extraction, training, and recognition components of the system are all designed to be script independent. The HMM training and recognition components are based on our Byblos continuous speech recognition system. The HMM parameters are estimated automatically from the training data, without the need for laborious hand-written rules. The system does not require pre-segmentation of the data, neither at the word level nor at the character level. Thus, the system can handle languages with cursive handwritten scripts in a straightforward manner. The script independence of the system is demonstrated with experimental results in three scripts that exhibit significant differences in glyph characteristics: English, Chinese, and Arabic. Results from an initial set of experiments are presented to demonstrate the viability of the proposed methodology.

Improvements in BBN's HMM-Based Offline Arabic Handwriting Recognition System

Offline handwriting recognition of free-flowing Arabic text is a challenging task due to the plethora of factors that contribute to the variability in the data. In this paper, we address some of these sources of variability, and present experimental results on a large corpus of handwritten documents. Specific techniques such as the application of context-dependent Hidden Markov Models (HMMs) for the cursive Arabic script, unsupervised adaptation to account for the stylistic variations across scribes, and image pre-processing to remove ruled-lines are explored. In particular, we proposed a novel integration of structural features in the HMM framework which exclusively results in a 9% relative improvement in performance. Overall, we demonstrate a relative reduction of 17% in word error rate over our baseline Arabic handwriting recognition system.

Improvements in hidden Markov model based Arabic OCR

This paper describes recent advances in hidden Markov model (HMM) based OCR for machine-printed arabic documents. A combination of script-independent and script-specific techniques are applied to glyph models and language models (LM). Script-independent techniques we applied are higher order n-gram LMs for N-best rescoring and discriminative estimation of glyph HMMs. Arabic specific techniques include the use of context-dependent HMMs for glyph modeling and Parts-of-Arabic-Words in language modeling. We present experimental results that demonstrate a 40% relative reduction in word error rate over the baseline configuration on a corpus of machine-printed Arabic documents.

Finding structure in noisy text: topic classification and unsupervised clustering

This paper addresses two types of classification of noisy, unstructured text such as newsgroup messages: (1) spotting messages containing topics of interest, and (2) automatic conceptual organization of messages without prior knowledge of topics of interest. In addition to applying our hidden Markov model methodology to spotting topics of interest in newsgroup messages, we present a robust methodology for rejecting messages which are off-topic. We describe a novel approach for automatically organizing a large, unstructured collection of messages. The approach applies an unsupervised topic clustering procedure to generate a hierarchical tree of topics.

Recent improvements and performance analysis of ASR and MT in a speech-to-speech translation system

We report on recent ASR and MT work on our English/Iraqi Arabic speech-to-speech translation system. We present detailed results for both objective and subjective evaluations of translation quality, along with a detailed analysis and categorization of translation errors. We also present novel ideas for quantifying the relative importance of different subjective error categories, and for assigning the blame for an error to a particular phrase pair in the translation model.

BBN TransTalk: Robust multilingual two-way speech-to-speech translation for mobile platforms

In this paper we present a speech-to-speech (S2S) translation system called the BBN TransTalk that enables two-way communication between speakers of English and speakers who do not understand or speak English. The BBN TransTalk has been configured for several languages including Iraqi Arabic, Pashto, Dari, Farsi, Malay, Indonesian, and Levantine Arabic. We describe the key components of our system: automatic speech recognition (ASR), machine translation (MT), text-to-speech (TTS), dialog manager, and the user interface (UI). In addition, we present novel techniques for overcoming specific challenges in developing high-performing S2S systems. For ASR, we present techniques for dealing with lack of pronunciation and linguistic resources and effective modeling of ambiguity in pronunciations of words in these languages. For MT, we describe techniques for dealing with data sparsity as well as modeling context. We also present and compare different user confirmation techniques for detecting errors that can cause the dialog to drift or stall.

Unsupervised HMM Adaptation Using Page Style Clustering

In this paper we present an innovative two-stage adaptation approach for handwriting recognition that is based on clustering of similar pages in the training data. In our approach, we first perform page clustering on training data using features such as contour slope, pen pressure, writing velocity, and stroke sparseness. Next, we adapt the writer-independent Hidden Markov models (HMMs) to each cluster in the training data. While decoding a test page, we first determine the cluster the test page belongs to and then decode the page with the model associated with that cluster. Experimental results with the two-stage adaptation show significant gains on a held-out validation set.

Recent improvements in BBN's English/Iraqi speech-to-speech translation system

We report on recent improvements in our English/Iraqi Arabic speech-to-speech translation system. User interface improvements include a novel parallel approach to user confirmation which makes confirmation cost-free in terms of dialog duration. Automatic speech recognition improvements include the incorporation of state-of-the-art techniques in feature transformation and discriminative training. Machine translation improvements include a novel combination of multiple alignments derived from various pre-processing techniques, such as Arabic segmentation and English word compounding, higher order N-grams for target language model, and use of context in form of semantic classes and part-of-speech tags.

Real-Time Speech-to-Speech Translation for PDAs

In this paper we present a speech-to-speech translation system configured for translingual communication in English and colloquial Iraqi on a mobile, handheld device. The end-to-end system employs a medium/large vocabulary n-gram speech recognition engine for recognizing English and colloquial Iraqi, a question canonicalizer for mapping a recognized English question or command to one of the questions supported in the system, a concept translation engine for translating recognized Iraqi text, and a text-to-speech synthesis engine for playing back the English translation for the Iraqi to the English speaker. In addition to describing the system architecture and the functionality of the components, we present optimization techniques that enable low-latency, real-time speech recognition on low-power hardware platforms.

Semantic translation error rate for evaluating translation systems

In this paper, we introduce a new metric which we call the semantic translation error rate, or STER, for evaluating the performance of machine translation systems. STER is based on the previously published translation error rate (TER) (Snover et al., 2006) and METEOR (Banerjee and Lavie, 2005) metrics. Specifically, STER extends TER in two ways: first, by incorporating word equivalence measures (WordNet and Porter stemming) standardly used by METEOR, and second, by disallowing alignments of concept words to non-concept words (aka stop words). We show how these features make STER alignments better suited for human-driven analysis than standard TER. We also present experimental results that show that STER is better correlated to human judgments than TER. Finally, we compare STER to METEOR, and illustrate that METEOR scores computed using the STER alignments have similar statistical properties to METEOR scores computed using METEOR alignments.