Marcos Calvo

Improvements on automatic speech segmentation at the phonetic level

In this paper, we present some recent improvements in our automatic speech segmentation system, which only needs the speech signal and the phonetic sequence of each sentence of a corpus to be trained. It estimates a GMM by using all the sentences of the training subcorpus, where each Gaussian distribution represents an acoustic class, which probability densities are combined with a set of conditional probabilities in order to estimate the probability densities of the states of each phonetic unit. The initial values of the conditional probabilities are obtained by using a segmentation of each sentence assigning the same number of frames to each phonetic unit. A DTW algorithm fixes the phonetic boundaries using the known phonetic sequence. This DTW is a step inside an iterative process which aims to segment the corpus and re-estimate the conditional probabilities. The results presented here demonstrate that the system has a good capacity to learn how to identify the phonetic boundaries.

A Multilingual SLU System Based on Semantic Decoding of Graphs of Words

In this paper, we present a statistical approach to Language Understanding that allows to avoid the effort of obtaining new semantic models when changing the language. This way, it is not necessary to acquire and label new training corpora in the new language. Our approach consists of learning all the semantic models in a target language and to do the semantic decoding of the sentences pronounced in the source language after a translation process. In order to deal with the errors and the lack of coverage of the translations, a mechanism to generalize the result of several translators is proposed. The graph of words generated in this phase is the input to the semantic decoding algorithm specifically designed to combine statistical models and graphs of words. Some experiments that show the good behavior of the proposed approach are also presented.

Multilingual Spoken Language Understanding using graphs and multiple translations

HighlightsTest-on-source multilingual speech understanding.Construction of graphs of words from multiple translations.Semantic decoding of graphs of words using statistical models.Unsupervised portability for Spoken Language Understanding. In this paper, we present an approach to multilingual Spoken Language Understanding based on a process of generalization of multiple translations, followed by a specific methodology to perform a semantic parsing of these combined translations. A statistical semantic model, which is learned from a segmented and labeled corpus, is used to represent the semantics of the task in a language. Our goal is to allow the users to interact with the system using other languages different from the one used to train the semantic models, avoiding the cost of segmenting and labeling a training corpus for each language. In order to reduce the effect of translation errors and to increase the coverage, we propose an algorithm to generate graphs of words from different translations. We also propose an algorithm to parse graphs of words with the statistical semantic model. The experimental results confirm the good behavior of this approach using French and English as input languages in a spoken language understanding task that was developed for Spanish.

Exploiting Multiple ASR Outputs for a Spoken Language Understanding Task

In this paper, we present an approach to Spoken Language Understanding, where the input to the semantic decoding process is a composition of multiple hypotheses provided by the Automatic Speech Recognition module. This way, the semantic constraints can be applied not only to a unique hypothesis, but also to other hypotheses that could represent a better recognition of the utterance. To do this, we have developed an algorithm to combine multiple sentences into a weighted graph of words, which is the input to the semantic decoding process. It has also been necessary to develop a specific algorithm to process these graphs of words according to the statistical models that represent the semantics of the task. This approach has been evaluated in a SLU task in Spanish. Results, considering different configurations of ASR outputs, show the better behavior of the system when a combination of hypotheses is considered.

A Phonetic-Based Approach to Query-by-Example Spoken Term Detection

Query-by-Example Spoken Term Detection QbE-STD tasks are usually addressed by representing speech signals as a sequence of feature vectors by means of a parametrization step, and then using a pattern matching technique to find the candidate detections. In this paper, we propose a phoneme-based approach in which the acoustic frames are first converted into vectors representing the a posteriori probabilities for every phoneme. This strategy is specially useful when the language of the task is a priori known. Then, we show how this representation can be used for QbE-STD using both a Segmental Dynamic Time Warping algorithm and a graph-based method. The proposed approach has been evaluated with a QbE-STD task in Spanish, and the results show that it can be an adequate strategy for tackling this kind of problems.

Cross-language transfer of semantic annotation via targeted crowdsourcing: task design and evaluation

AbstractModern data-driven spoken language systems (SLS) require manual semantic annotation for training spoken language understanding parsers. Multilingual porting of SLS demands significant manual effort and language resources, as this manual annotation has to be replicated. Crowdsourcing is an accessible and cost-effective alternative to traditional methods of collecting and annotating data. The application of crowdsourcing to simple tasks has been well investigated. However, complex tasks, like cross-language semantic annotation transfer, may generate low judgment agreement and/or poor performance. The most serious issue in cross-language porting is the absence of reference annotations in the target language; thus, crowd quality control and the evaluation of the collected annotations is difficult. In this paper we investigate targeted crowdsourcing for semantic annotation transfer that delegates to crowds a complex task such as segmenting and labeling of concepts taken from a domain ontology; and evaluation using source language annotation. To test the applicability and effectiveness of the crowdsourced annotation transfer we have considered the case of close and distant language pairs: Italian–Spanish and Italian–Greek. The corpora annotated via crowdsourcing are evaluated against source and target language expert annotations. We demonstrate that the two evaluation references (source and target) highly correlate with each other; thus, drastically reduce the need for the target language reference annotations.

Combining Several ASR Outputs in a Graph-Based SLU System

In this paper, we present an approach to Spoken Language Understanding (SLU) where we perform a combination of multiple hypotheses from several Automatic Speech Recognizers (ASRs) in order to reduce the impact of recognition errors in the SLU module. This combination is performed using a Grammatical Inference algorithm that provides a generalization of the input sentences by means of a weighted graph of words. We have also developed a specific SLU algorithm that is able to process these graphs of words according to a stochastic semantic modelling.The results show that the combinations of several hypotheses from the ASR module outperform the results obtained by taking just the 1-best transcription.

Voice-QA: Evaluating the Impact of Misrecognized Words on Passage Retrieval

Question Answering is an Information Retrieval task where the query is posed using natural language and the expected result is a concise answer. Voice-activated Question Answering systems represent an interesting application, where the question is formulated by speech. In these systems, an Automatic Speech Recognition module can be used to transcribe the question. Thus, recognition errors may be introduced, producing a significant effect on the answer retrieval process. In this work we study the relationship between some features of misrecognized words and the retrieval results. The features considered are the redundancy of a word in the result set and its inverse document frequency calculated over the collection. The results show that the redundancy of a word may be an important clue on whether an error on it would deteriorate the retrieval results, at least if a closed model is used for speech recognition.