Automated Teller Machine (ATM) service providers are increasingly challenged with improving the quality of customer service while reducing the cost of cash flow management. Effectively balancing the need to have enough cash in the ATMs to avoid out-of-cash incidents as well as to reduce the cash interest cost and the cash refill cost challenges the most experienced cash flow management teams. In this paper we propose an optimization framework for managing the ATM cash flow network. The interactions among various constraints and cost factors are included in the framework to allow decision-making regarding the optimal cash refill amount and schedule. We demonstrate the effectiveness of the proposed approach using sample data from a large commercial bank.

Ticket annotation and search has become an essential research subject for the successful delivery of IT operational analytics. Millions of tickets are created yearly to address business users' IT related problems. In IT service desk management, it is critical to first capture the pain points for a group of tickets to determine root cause; secondly, to obtain the respective distributions in order to layout the priority of addressing these pain points. An advanced ticket analytics system utilizes a combination of topic modeling, clustering and Information Retrieval (IR) technologies to address the above issues and the corresponding architecture which integrates of these features will allow for a wider distribution of this technology and progress to a significant financial benefit for the system owner. Topic modeling has been used to extract topics from given documents; in general, each topic is represented by a unigram language model. However, it is not clear how to interpret the results in an easily readable/understandable way until now. Due to the inefficiency to render top concepts using existing techniques, in this paper, we propose a probabilistic framework, which consists of language modeling (especially the topic models), Part-Of-Speech (POS) tags, query expansion, retrieval modeling and so on for the practical challenge. The rigorously empirical experiments demonstrate the consistent and utility performance of the proposed method on real datasets.

Extractive text or speech summarization manages to select a set of salient sentences from an original document and concatenate them to form a summary, enabling users to better browse through and understand the content of the document. A recent stream of research on extractive summarization is to employ the language modeling (LM) approach for important sentence selection, which has proven to be effective for performing speech summarization in an unsupervised fashion. However, one of the major challenges facing the LM approach is how to formulate the sentence models and accurately estimate their parameters for each sentence in the document to be summarized. In view of this, our work in this paper explores a novel use of recurrent neural network language modeling (RNNLM) framework for extractive broadcast news summarization. On top of such a framework, the deduced sentence models are able to render not only word usage cues but also long-span structural information of word co-occurrence relationships within broadcast news documents, getting around the need for the strict bag-of-words assumption. Furthermore, different model complexities and combinations are extensively analyzed and compared. Experimental results demonstrate the performance merits of our summarization methods when compared to several well-studied state-of-the-art unsupervised methods.

Statistical language modeling (LM) that purports to quantify the acceptability of a given piece of text has long been an interesting yet challenging research area.In particular, language modeling for information retrieval (IR) has enjoyed remarkable empirical success; one emerging stream of the LM approach for IR is to employ the pseudo-relevance feedback process to enhance the representation of an input query so as to improve retrieval effectiveness.This paper presents a continuation of such a general line of research and the main contribution is threefold.First, we propose a principled framework which can unify the relationships among several widely-used query modeling formulations.Second, on top of the successfully developed framework, we propose an extended query modeling formulation by incorporating critical query-specific information cues to guide the model estimation.Third, we further adopt and formalize such a framework to the speech recognition and summarization tasks.A series of empirical experiments reveal the feasibility of such an LM framework and the performance merits of the deduced models on these two tasks.

A part and parcel of any automatic speech recognition (ASR) system is language modeling (LM), which helps to constrain the acoustic analysis, guide the search through multiple candidate word strings, and quantify the acceptability of the final output hypothesis given an input utterance. Despite the fact that the n-gram model remains the predominant one, a number of novel and ingenious LM methods have been developed to complement or be used in place of the n-gram model. A more recent line of research is to leverage information cues gleaned from pseudo-relevance feedback (PRF) to derive an utterance-regularized language model for complementing the n-gram model. This paper presents a continuation of this general line of research and its main contribution is two-fold. First, we explore an alternative and more efficient formulation to construct such an utterance-regularized language model for ASR. Second, the utilities of various utterance-regularized language models are analyzed and compared extensively. Empirical experiments on a large vocabulary continuous speech recognition (LVCSR) task demonstrate that our proposed language models can offer substantial improvements over the baseline n-gram system, and achieve performance competitive to, or better than, some state-of-the-art language models.

Query-by-example information retrieval provides users a flexible but efficient way to accurately describe their information needs. The query exemplars are usually long and in the form of either a partial or even a full document. However, they may contain extraneous terms that would have potential negative impacts on the retrieval performance. In order to alleviate those negative impacts, we propose a novel term-based query reduction mechanism so as to improve the informativeness of verbose query exemplars. We also explore the notion of term discrimination power to select a salient subset of query terms automatically. Experiments on the TDT Chinese collection show that the proposed approach is indeed effective and promising.

Natural language call routing systems are an attractive alternative to interactive voice response systems and directed dialog systems for automating customer service functions. However, the up-front development cost of these systems is an obstacle to their widespread adoption. Much of the cost is associated with the collection and annotation of development data that are used in initial system construction. In this work, we show how the statistical language model and action classifier needed for speech action classification can be developed for a customer's call routing application using no development data. On live data, our approach has comparable performance to a model trained using 100k utterances of in-domain development data. Furthermore, our approach handles the “unknown” class more robustly. These promising experimental results indicate that our method can be used to rapidly and inexpensively deploy call routing systems.

This paper addresses parallel data extraction from the quasi–parallel corpora generated in a crowd-sourcing project where ordinary people watch tv shows and movies and transcribe/translate what they hear, creating document pools in different languages. Since they do not have guidelines for naming and performing translations, it is often not clear which documents are the translations of the same show/movie and which sentences are the translations of the each other in a given document pair. We introduce a method for automatically pairing documents in two languages and extracting parallel sentences from the paired documents. The method consists of three steps: i) document pairing, ii) sentence pair alignment of the paired documents, and iii) context extrapolation to boost the sentence pair coverage. Human evaluation of the extracted data shows that 95 % of the extracted sentences carry useful information for translation. Experimental results also show that using the extracted data provides significant gains over the baseline statistical machine translation system built with manually annotated data. Index Terms: data extraction, comparable data, machine translation

ABSTRACTIn this paper we describe some experiments on the Aurora 2 noisydigits database. The algorithms that we used can be broadly clas-sified into noise robustness techniques based on a linear-channelmodel of the acoustic environment such as CDCN [1] and its novelvariant termed Alignment-based CDCN( ACDCN , proposed here),and techniques which do not assume any particular knowledgeabout thestructure of the environment or noise conditions affectingthe speech signal such as discriminant feature space transforma-tions and speaker/channel adaptation. We present recognition ex-periments for both the clean training data and the multi-conditiontraining data scenarios.1. INTRODUCTIONIn this paper we describe the system and techniques for the Aurora2 noisy digits database and the results obtained. We developed twosets of acoustic models: the first set of models was trained on cleandata only and the second set was trained on the multi-conditiontraining data. For the system trained on clean data only, we appliedCDCN [1] and a novel variant of this technique called Alignment-based CDCN (

Two discriminant measures for HMM states to improve the effectiveness on HMM training are presented. In HMM based speech recognition, the context-dependent states are usually modeled by Gaussian mixture distributions. In general, the number of Gaussian mixtures for each state is fixed or proportional to the amount of training data. From our study, some of the states are "non-aggressive" compared to others, and a higher acoustic resolution is required for them. Two methods are presented in this paper to determine those non-aggressive states. The first approach uses the recognition accuracy of the states and the second method is based on a rank distribution of states. Baseline systems, trained by a fixed number of Gaussian mixtures for each state, having 33 K and 120 K Gaussians, yield 14.57% and 13.04% word error rates, respectively. Using our approach, a 38 K Gaussian system was constructed that reduces the error rate to 13.95%. The average ranks of non-aggressive states in rank lists of testing data were also seen to dramatic improve compared to the baseline systems.

We describe our development work on a telephonebased conversational system in the domain of mutual fund transactions. This system uses several components including robust large vocabulary continuous speech recognition, natural language understanding, dialog management, and text-to-speech synthesis technologies.

Large vocabulary automatic speech recognition might assist hearing impaired telephone users by displaying a transcription of the incoming side of the conversation, but the system would have to achieve su cient accuracy on conversationalstyle, telephone-bandwidth speech. We describe our development work toward such a system. This work comprised three phases: Experiments with clean data ltered to 200-3500Hz, experiments with real telephone data, and language model development. In the rst phase, the speaker independent error rate was reduced from 25% to 12% by using MLLT, increasing the number of cepstral components from 9 to 13, and increasing the number of Gaussians from 30,000 to 120,000. The resulting system, however, performed less well on actual telephony, producing an error rate of 28.4%. By additional adaptation and the use of an LDA and CDCN combination, the error rate was reduced to 19.1%. Speaker adaptation reduces the error rate to 10.96%. These results were obtained with read speech. To explore the language-model requirements in a more realistic situation, we collected some conversational speech with an arrangement in which one participant could not hear the conversation but only saw recognizer output on a screen. We found that a mixture of language models, one derived from the Switchboard corpus and the other from prepared texts, resulted in approximately 10% fewer errors than either model alone.

This report describes microphone arrays and parallel signal processing for high-quality sound capture in noisy, reverberant enclosures. The technique incorporates matched-filtering of individual sensors and parallel processing to provide spatial volume selectivity that mitigates effects of noise interference and multipath distortion. Truncated causal approximations to ideal matched-filters and arrays composed of randomly distributed transducers reduce complexity and improve the quality of sound capture. The method outperforms traditional delay-and-sum beamformers which provide only directional selectivity.

An extensive speech database was collected in two experimental enclosures for research on microphone arrays. The first enclosure is a digitally controlled variable acoustics facility where the reverberation time can be adjusted from 1.7 s to 0.1 s, and the second is a regular hard-walled laboratory. Two different sets of data were collected to address source location and spatial volume selectivity issues. Preliminary results on Matched-filter processing and on Source location are reported for the new database. I. Introduction Performance of microphone sound pick up is degraded by deleterious properties of the acoustic environment, such as multipath distortion (reverberation) and ambient noise. The degradation becomes more prominent in a teleconferencing environment in which the microphone is positioned far away from the speaker. Besides, the ideal teleconference should feel as easy and natural as face-to-face communication with another person. This suggests hands-free sound capture w...

Performance of a delay-and-sum beamformer is typically degraded in a reverberant enclosure because the beam captures not only the desired source signal (direct path) but also all images along the beam axis. Matched-filter processing of microphone arrays is shown to improve the quality of sound capture in reverberant environments. This paper also reports the performance of matched-filter array processing in a real room. Significant improvements are shown over single microphones and the traditional delay-and-sum beamformer.

Hands-free operation of speech processing equipment is sometimes desired so that the user is unencumbered by hand-held or body-worn microphones. This paper explores the use of array microphones to capture speech under adverse acoustic conditions, and provide input to a system for automatic speaker identification. The system is evaluated using reverberated speech signals, generated by a computer model of room acoustics and transduced by different simulated microphone-arrays. For comparison, the system is also evaluated using close-talking microphone input. It is found that 2-D matched-filter microphone arrays are capable of producing high speaker identification scores in a hostile acoustic environment, such as multipath distortion and competing noise sources. The paper also explores the influence of vector quantization techniques, codebook size, and order of cepstrum coefficients on the performance of the speaker identification system.>