This paper develops the MUFIN technique for extreme classification (XC) tasks with millions of labels where data-points and labels are endowed with visual and textual de-scriptors. Applications of MUFIN to product-to-product recommendation and bid query prediction over several mil-lions of products are presented. Contemporary multi-modal methods frequently rely on purely embedding-based meth-ods. On the other hand, XC methods utilize classifier ar-chitectures to offer superior accuracies than embedding-only methods but mostly focus on text-based categorization tasks. MUFIN bridges this gap by reformulating multi-modal categorization as an XC problem with several mil-lions of labels. This presents the twin challenges of devel-oping multi-modal architectures that can offer embeddings sufficiently expressive to allow accurate categorization over millions of labels; and training and inference routines that scale logarithmically in the number of labels. MUFIN de-velops an architecture based on cross-modal attention and trains it in a modular fashion using pre-training and positive and negative mining. A novel product-to-product rec-ommendation dataset MM-AmazonTitles-300K containing over 300K products was curated from publicly available amazon.com listings with each product endowed with a title and multiple images. On the MM-AmazonTitles-300K and Polyvore datasets, and a dataset with over 4 million labels curated from click logs of the Bing search engine, MUFIN offered at least 3% higher accuracy than leading text-based, image-based and multi-modal techniques.

Non-Sentential Utterances (NSUs) are short utterances that do not have the form of a full sentence but nevertheless convey a complete sentential meaning in the context of a conversation.NSUs are frequently used to ask follow up questions during interactions with question answer (QA) systems resulting into in-correct answers being presented to their users.Most of the current methods for resolving such NSUs have adopted rule or grammar based approach and have limited applicability.In this paper, we present a data driven statistical method for resolving such NSUs.Our method is based on the observation that humans identify keyword appearing in an NSU and place them in the context of conversation to construct a meaningful sentence.We adapt the keyword to question (K2Q) framework to generate natural language questions using keywords appearing in an NSU and its context.The resulting questions are ranked using different scoring methods in a statistical framework.Our evaluation on a data-set collected using mTurk shows that the proposed method perform significantly better than the previous work that has largely been rule based.

In this work we propose an unsupervised framework to construct a shallow domain ontology from corpus. It is essential for Information Retrieval systems, Question-Answering systems, Dialogue etc. to identify important concepts in the domain and the relationship between them. We identify important domain terms of which multi-words form an important component. We show that the incorporation of multi-words improves parser performance, resulting in better parser output, which improves the performance of an existing Question-Answering system by upto 7%. On manually annotated smartphone dataset, the proposed system identifies 40:87% of the domain terms, compared to 22% recall obtained using WordNet, 43:77% by Yago and 53:74% by BabelNet respectively. However, it does not use any manually annotated resource like the compared systems. Thereafter, we propose a framework to construct a shallow ontology from the discovered domain terms by identifying four domain relations namely, Synonyms ('similar-to'), Type-Of ('is-a'), Action-On ('methods') and Feature-Of ('attributes'), where we achieve significant performance improvement over WordNet, BabelNet and Yago without using any mode of supervision or manual annotation.

In a customer support scenario, a lot of valuable information is recorded in the form of `case logs'. Case logs are primarily written for future references or manual inspections and therefore are written in a hasty manner and are very noisy. In this paper, we propose techniques that exploit these case logs to mine real customer concerns or problems and then map them to well written knowledge articles for that enterprise. This mapping results into generation of question-answer (QA) pairs. These QA pairs can be used for a variety of applications such as dynamically updating the frequently-asked-questions (FAQs), updating the knowledge repository etc. In this paper we show the utility of these discovered QA pairs as training data for a question-answering system. Our approach for mining the case logs is based on a composite model consisting of two generative models, viz, hidden Markov model (HMM) and latent Dirichlet allocation (LDA) model. The LDA model explains the long-range dependencies across words due to their semantic similarity and HMM models the sequential patterns present in these case logs. Such processing results in crisp `problem statement' segments which are indicative of the real customer concerns. Our experiments show that this approach finds crisp problem-statements in 56% of the cases and outperforms other alternate methods for segmentation such as HMM, LDA and conditional random field (CRF). After finding these crisp problem-statements, appropriate answers are looked up from an existing knowledge repository index forming candidate QA pairs. We show that considering only the problemstatement segments for which the answers can be found further improves the segmentation performance to 82%. Finally, we show that when these QA pairs are used as training data, the performance of a question-answering system can be improved significantly.

The hidden Markov model (HMM) is widely popular as the de facto tool for representing temporal data; in this paper, we add to its utility in the sequence clustering domain - we describe a novel approach that allows us to directly control purity in HMM-based clustering algorithms. We show that encouraging sparsity in the observation probabilities increases cluster purity and derive an algorithm based on lpregularization; as a corollary, we also provide a different and useful interpretation of the value of p in Renyi p-entropy. We test our method on the problem of clustering non-speech audio events from the BBC sound effects corpus. Experimental results confirm that our approach does learn purer clusters, with (unweighted) average purity as high as 0.88 - a considerable improvement over both the baseline HMM (0.72) and k-means clustering (0.69).

The social Customer Relationship Management (CRM) landscape is attracting significant attention from customers and enterprises alike as a sustainable channel for tracking, managing and improving customer relations. Enterprises are taking a hard look at this open, unmediated platform because the community effect generated on this channel can have a telling effect on their brand image, potential market opportunity and customer loyalty. In this work we present our experiences in building a system that mines conversations on social platforms to identify and prioritize those posts and messages that are relevant to enterprises. The system presented in this work aims to empower an agent or a representative in an enterprise to monitor, track and respond to customer communication while also encouraging community participation.

Word clouds are extensively used to present a summary of the prominent words in a document on the World Wide Web. Such clouds give the user an idea about the content of the document. In this paper we present a mechanism to create and render an audio cloud for audio content. Such audio clouds are expected to provide a similar summary of the audio documents. They have wide applicability in various domains, especially for low-literate users who currently do not use the Internet but interact with audio-based systems.

In this paper, we deal with the problem of predicting how much attention a newly submitted post would receive from fellow community members of closed communities in social networking sites. Though the concept of attention is subjective, the number of comments received by a post serves as a very good indicator of the same. Unlike previous work which primarily made use of either content features or the network features (friendship links on the network), we exploit both the content features and community level features (for instance, what time of the day is the community more active) for tackling this problem. Further, we focus on dedicated pages of corporate brands on social media websites and accordingly extract important features from the content and community activity of such brand pages. The attention prediction task finds direct application in the listening, monitoring and engaging activities of the businesses that have such brand-pages. In this paper, we formulate the problem of attention prediction on social media brand pages.

With evolving Web, short length parallel corpora is becoming very common and some of these include user queries, web snippets etc. This paper concerns situations where short length parallel corpora has to be analyzed in order to find meaningful unit-alignment. This is similar to dealing with parallel corpora where a sentence level alignment of translations is required, but differs in that the alignment is to be inferred at unit (word or phrase) level. A Conditional Random Field (CRF) based approach is proposed to discover this unit alignment. Given pairs of semantically or syntactically similar entities, the problem is formulated as that of mutual segmentation and sequence alignment problem. The mutual segmentation refers to the process of segmenting the first entity based on units (or labels) in the second entity and vice-versa. The process of optimizing this mutual segmentation also results in optimal unit alignment. Since our training data is not segmented and unit-aligned, we modify the CRF objective function to accommodate unsupervised data and iterative learning. We have applied this framework to Web Search domain and specifically for query reformulation task. Finally, our experiments suggest that the proposed approach indeed results in meaningful alternatives of the original query.

This paper presents recent developments at our site toward speech recognition using decision tree based acoustic models. Previously, robust decision trees have been shown to achieve better performance compared to standard Gaussian mixture model (GMM) acoustic models. This was achieved by converting hard questions (decisions) of a standard tree into soft questions using sigmoid function. In this paper, we report our work where soft-decision trees are trained from scratch. These softdecision trees are shown to yield better speech recognition accuracy compared to standard GMM acoustic models on Aurora digit recognition task.

This paper deals with the task of detection of a given keyword in continuous speech. We build upon a previously proposed algorithm where a modified Viterbi search algorithm is used to detect keywords, without requiring any explicit garbage or filler models. In this work, the concept of durational entropy is used to further discard a large fraction of false alarm errors. Durational entropy is defined as the entropy of the distribution of state occupancies. A method to recursively compute it for all Viterbi paths is also presented in this paper. Experimental results on one hour of broadcast news data suggest that durational entropy constraints can indeed be used to avoid a large number of false alarms errors at a minimal cost of degradation in keyword detection accuracy.

This paper presents a comparative study of four different approaches to automatic age and gender classification using seven classes on a telephony speech task and also compares the results with human performance on the same data. The automatic approaches compared are based on (1) a parallel phone recognizer, derived from an automatic language identification system; (2) a system using dynamic Bayesian networks to combine several prosodic features; (3) a system based solely on linear prediction analysis; and (4) Gaussian mixture models based on MFCCs for separate recognition of age and gender. On average, the parallel phone recognizer performs as well as Human listeners do, while loosing performance on short utterances. The system based on prosodic features however shows very little dependence on the length of the utterance.

In multi-party teleconferencing, the transport of separate speech streams to a particular user and the subsequent spatial rendering of the different streams enables a more efficient communication. A simple means of spatial presentation at client side is that of binaural rendering and headphone presentation. For downwardcompatibility, e.g. when the transport mechanism does not support multiple parallel downlink streams, a system is proposed that combines an automatic speaker classification mechanism with a spatial rendering of the segregated streams. The combined system aims at a better separability of the speakers than conventional systems. The paper details the two basic components, namely automatic speaker classification, and binaural rendering. Based on a first evaluation of the approach, a proof of concept is provided, and directions for further improvement are discussed.

Anger detection is a topic that is gaining more and more attention with voice portal carriers, as it can be useful for quality measurement and emotion-aware dialog strategies. In the context of a prototype voice portal we describe methods to search for training data, report on the performance of the prosodic classifier under real world conditions and explore the use of dialog information for anger prediction. The results show that, although significantly worse than under laboratory conditions, anger detection still works well above chance level and can be used to enhance real world voice-portal usability. Index Terms: Emotion Recognition, Voice Portal, Speech Classification, Dialogue System.

The paper presents a new approach toward automatic annotation of meetings in terms of speaker identities and their locations. This is achieved by segmenting the audio recordings using two independent sources of information: magnitude spectrum analysis and sound source localization. We combine the two in an appropriate HMM framework. There are three main advantages of this approach. First, it is completely unsupervised, i.e. speaker identities and number of speakers and locations are automatically inferred. Second, it is threshold-free, i.e. the decisions are made without the need of a threshold value which generally requires an additional development dataset. The third advantage is that the joint segmentation improves over the speaker segmentation derived using only acoustic features. Experiments on a series of meetings recorded in the IDIAP smart meeting room demonstrate the effectiveness of this approach.

This paper presents overview of an online audio indexing system, which creates a searchable index of speech content embedded in digitized audio files. This system is based on our recently proposed offline audio segmentation techniques. As the data arrives continuously, the system first finds boundaries of the acoustically homogenous segments. Next, each of these segments is classified as speech, music or {\\it mixture} classes, where mixtures are defined as regions where speech and other non-speech sounds are present simultaneously and noticeably. The speech segments are then clustered together to provide consistent speaker labels. The speech and mixture segments are converted to text via an ASR system. The resulting words are time-stamped together with other metadata information (speaker identity, speech confidence score) in an XML file to rapidly identify and access target segments. In this paper, we analyze the performance at each stage of this audio indexing system and also compare it with the performance of the corresponding offline modules.

Most commonly used criteria for speaker change detection like log likelihood ratio (LLR) and Bayesian information criterion (BIC) have an adjustable threshold/penalty parameter to make speaker change decisions. These parameters are not always robust to different acoustic conditions and have to be tuned. In this letter, we present a criterion which can be used to identify speaker changes in an audio stream without such tuning. The criterion consists of calculating the LLR of two models with the same number of parameters. Results on the Hub4 1997 evaluation set indicate that we achieve a performance comparable to using BIC with optimal penalty term.

In this paper we present a new approach towards high performance speech/music segmentation on realistic tasks related to the automatic transcription of broadcast news. In the approach presented here, the local probability density function (PDF) estimators trained on clean microphone speech are used as a channel model at the output of which the entropy and “dynamism” will be measured and integrated over time through a 2-state (speech and and non-speech) hidden Markov model (HMM) with minimum duration constraints. The parameters of the HMM are trained using the EM algorithm in a completely unsupervised manner. Different experiments, including a variety of speech and music styles, as well as different segment durations of speech and music signals (real data distribution, mostly speech, or mostly music), will illustrate the robustness of the approach, which in each case achieves a frame-level accuracy greater than 94%.

An HMM-based speaker clustering framework is presented, where the number of speakers and segmentation boundaries are unknown \\emph{a priori}. Ideally, the system aims to create one pure cluster for each speaker. The HMM is ergodic in nature with a minimum duration topology. The final number of clusters is determined automatically by merging closest clusters and retraining this new cluster, until a decrease in likelihood is observed. In the same framework, we also examine the effect of using only the features from highly voiced frames as a means of improving the robustness and computational complexity of the algorithm. The proposed system is assessed on the 1996 HUB-4 evaluation test set in terms of both cluster and speaker purity. It is shown that the number of clusters found often correspond to the actual number of speakers.