We present a novel approach for improving communication success between users of speech-to-speech translation systems by automatically detecting errors in the output of automatic speech recognition (ASR) and statistical machine translation (SMT) systems. Our approach initiates system-driven targeted clarification about errorful regions in user input and repairs them given user responses. Our system has been evaluated by unbiased subjects in live mode, and results show improved success of communication between users of the system.

Statistical machine translation (SMT) systems for spoken languages suffer from conversational speech phenomena, in particular, the presence of speech disfluencies. We examine the impact of disfluencies from broadcast conversation data on our hierarchical phrase-based SMT system and implement automatic disfluency removal approaches for cleansing the MT input. We evaluate the efficacy of proposed approaches and investigate the impact of disfluency removal on SMT performance across different disfluency types. We show that for translating Mandarin broadcast conversational transcripts into English, our automatic disfluency removal approaches could produce significant improvement in BLEU and TER.

We describe our recent effort implementing SRI's UMPC-based Pashto speech-to-speech (S2S) translation system on a smart phone running the Android operating system. In order to maintain very low latencies of system response on computationally limited smart phone platforms, we developed efficient algorithms and data structures and optimized model sizes for various system components. Our current Android-based S2S system requires less than one-fourth the system memory and significantly lower processor speed with a sacrifice of 15% relative loss of system accuracy, compared to a laptop-based platform.

Recently, the use of syntax has very effectively improved machine translation (MT) quality in many text translation tasks. However, using syntax in speech translation poses additional challenges because of disfluencies and other spoken language phenomena, and of errors introduced by automatic speech recognition (ASR). In this paper, we investigate the effect of using syntax in a large-scale audio document translation task targeting broadcast news and broadcast conversations. We do so by comparing the performance of three synchronous context-free grammar based translation approaches: 1) hierarchical phrase-based translation, 2) syntax-augmented MT, and 3) string-to-dependency MT. The results show a positive effect of explicitly using syntax when translating broadcast news, but no benefit when translating broadcast conversations. The results indicate that improving the robustness of syntactic systems against conversational language style is important to their success and requires future effort. Index Terms: syntax, machine translation, audio document

We present our recent work on developing large-vocabulary Arabic-to-English and Chinese-to-English speech-to-text translation systems for the January 2008 Global Autonomous Language Exploitation (GALE) retest evaluation. Two audio genres were involved in the evaluation: broadcast news and broadcast conversation. Our system, following the hierarchical phrase-based translation approach, has a two-pass decoding strategy, with the first-pass integrated search generating 3000 unique n-best lists, which are then reranked by several different language models in the second pass. We emphasize our work on adapting the system, which was mostly trained on text data, to the speech genres, including number tokenization, punctuation compensation, and various optimization techniques. We present our results on several different tuning and testing data sets used for system development. Index Terms : speech-to-text translation, hierarchical-phrasebased translation, n-best reranking

Performance of statistical machine translation (SMT) systems relies on the availability of a large parallel corpus which is used to estimate translation probabilities. However, the generation of such corpus is a long and expensive process. In this paper, we introduce two methods for efficient selection of training data to be translated by humans. Our methods are motivated by active learning and aim to choose new data that adds maximal information to the currently available data pool. The first method uses a measure of disagreement between multiple SMT systems, whereas the second uses a perplexity criterion. We performed experiments on Chinese-English data in multiple domains and test sets. Our results show that we can select only one-fifth of the additional training data and achieve similar or better translation performance, compared to that of using all available data.

This paper presents an extensive evaluation of five different alignments and investigates their impact on the corresponding MT system output. We introduce new measures for intrinsic evaluations and examine the distribution of phrases and untranslated words during decoding to identify which characteristics of different alignments affect translation. We show that precision-oriented alignments yield better MT output (translating more words and using longer phrases) than recall-oriented alignments.

The frequent occurrence of divergences —structural differences between languages---presents a great challenge for statistical word-level alignment and machine translation. This paper describes the adaptation of DUSTer, a divergence unraveling package, to Hindi during the DARPA TIDES-2003 Surprise Language Exercise. We show that it is possible to port DUSTer to Hindi in under 3 days.

Site maps are essential to assist users in navigating a Web site. Most of the site maps are constructed manually and are static. However, di erent users may have di erent preferences and purposes for using a Web site. For example, a user may want to see a more detailed map while another user prefers a more abstract map. Two users looking for di erent topics at a large portal site would bene t more from two site maps with di erent focuses than a single map. In this paper, we present a technique for automatically constructing multi-granular and topic-focused site maps by utilizing directory paths, page contents, and link structures. In these site maps, the Web site topology is preserved and document importance, indicated by citation and semantic relevancy to user's topics of interest, is used for prioritizing the presentation of pages and directories. Experiments on real Web data have been conducted to validate the usefulness of the technique.

We present an efficient algorithm (UWEP) for updating large itemsets when new transactions are added to the set of old transactions.UWEP employs a dynamic lookahead strategy in updating the existing large itemsets by detecting and removing those that will no longer remain large after the contribution of the new set of transactions.It differs from the other update algorithms by scanning the existing database at most once and the new database exactly once.Moreover, it generates and counts the minimum number of candidates in the new database.The experiments on synthetic data show that UWEP outperforms the existing algorithms in terms of the candidates generated and counted.