This paper presents a description of our system for the Albayzin 2012 LRE competition. One of the main characteristics of this evaluation was the reduced number of available files for training the system, especially for the empty condition where no training data set was provided but only a development set. In addition, the whole database was created from online videos and around one third of the training data was labeled as noisy files. Our primary system was the fusion of three different i-vector based systems: one acoustic system based on MFCCs, a phonotactic system using trigrams of phone-posteriorgram counts, and another acoustic system based on RPLPs that improved robustness against noise. A contrastive system that included new features based on the glottal source was also presented. Official and post-evaluation results for all the conditions using the proposed metrics for the evaluation and the Cavgmetric are presented in the paper.

This paper presents the SAILSE Project (Sistema Avanzado de Información en Lengua de Signos Española ? Spanish Sign Language Advanced Information System). This project aims to develop an interactive system for facilitating the communication between a hearing and a deaf person. The first step has been the linguistic study, including a sentence collection, its translation into LSE (Lengua de Signos Española - Spanish Sign Language), and sign generation. After this analysis, the paper describes the interactive system that integrates an avatar to represent the signs, a text to speech converter and several translation technologies. Finally, this paper presents the set up carried out with deaf people and the main conclusions extracted from it.

Several methods to improve multiple distant microphone (MDM) speaker diarization based on Time Delay of Arrival (TDOA) features are evaluated in this paper. All of them avoid the use of a single reference channel to calculate the TDOA values and, based on different criteria, select among all possible pairs of microphones a set of pairs that will be used to estimate the TDOA's. The evaluated methods have been named the "Dynamic Margin" (DM), the "Extreme Regions" (ER), the "Most Common" (MC), the "Cross Correlation" (XCorr) and the "Principle Component Analysis" (PCA). It is shown that all methods improve the baseline results for the development set and four of them improve also the results for the evaluation set. Improvements of 3.49% and 10.77% DER relative are obtained for DM and ER respectively for the test set. The XCorr and PCA methods achieve an improvement of 36.72% and 30.82% DER relative for the test set. Moreover, the computational cost for the XCorr method is 20% less than the baseline.

Two new features have been proposed and used in the Rich Transcription Evaluation 2009 by the Universidad Politécnica de Madrid, which outperform the results of the baseline system. One of the features is the intensity channel contribution, a feature related to the location of the speaker. The second feature is the logarithm of the interpolated fundamental frequency. It is the first time that both features are applied to the clustering stage of multiple distant microphone meetings diarization. It is shown that the inclusion of both features improves the baseline results by 15.36% and 16.71% relative to the development set and the RT 09 set, respectively. If we consider speaker errors only, the relative improvement is 23% and 32.83% on the development set and the RT09 set, respectively.

This paper proposes the use of Factored Translation Models (FTMs) for improving a Speech into Sign Language Translation System. These FTMs allow incorporating syntactic-semantic information during the translation process. This new information permits to reduce significantly the translation error rate. This paper also analyses different alternatives for dealing with the non-relevant words. The speech into sign language translation system has been developed and evaluated in a specific application domain: the renewal of Identity Documents and Driver’s License. The translation system uses a phrase-based translation system (Moses). The evaluation results reveal that the BLEU (BiLingual Evaluation Understudy) has improved from 69.1% to 73.9% and the mSER (multiple references Sign Error Rate) has been reduced from 30.6% to 24.8%.

The time delay of arrival (TDOA) between multiple microphones has been used since 2006 as a source of information (localization) to complement the spectral features for speaker diarization. In this paper, we propose a new localization feature, the intensity channel contribution (ICC) based on the relative energy of the signal arriving at each channel compared to the sum of the energy of all the channels. We have demonstrated that by joining the ICC features and the TDOA features, the robustness of the localization features is improved and that the diarization error rate (DER) of the complete system (using localization and spectral features) has been reduced. By using this new localization feature, we have been able to achieve a 5.2% DER relative improvement in our development data, a 3.6% DER relative improvement in the RT07 evaluation data and a 7.9% DER relative improvement in the last year's RT09 evaluation data.

This paper describes the development of an Advanced Speech Communication System for Deaf People and its field evaluation in a real application domain: the renewal of Driver’s License. The system is composed of two modules. The first one is a Spanish into Spanish Sign Language (LSE: Lengua de Signos Española) translation module made up of a speech recognizer, a natural language translator (for converting a word sequence into a sequence of signs), and a 3D avatar animation module (for playing back the signs). The second module is a Spoken Spanish generator from sign writing composed of a visual interface (for specifying a sequence of signs), a language translator (for generating the sequence of words in Spanish), and finally, a text to speech converter. For language translation, the system integrates three technologies: an example based strategy, a rule based translation method and a statistical translator. This paper also includes a detailed description of the evaluation carried out in the Local Traffic Office in the city of Toledo (Spain) involving real government employees and deaf people. This evaluation includes objective measurements from the system and subjective information from questionnaires

This paper describes the development of a Spoken Spanish generator from sign-writing. The sign language considered was the Spanish sign language (LSE: Lengua de Signos Española). This system consists of an advanced visual interface (where a deaf person can specify a sequence of signs in sign-writing), a language translator (for generating the sequence of words in Spanish), and finally, a text to speech converter. The visual interface allows a sign sequence to be defined using several sign-writing alternatives. The paper details the process for designing the visual interface proposing solutions for HCI-specific challenges when working with the Deaf (i.e. important difficulties in writing Spanish or limited sign coverage for describing abstract or conceptual ideas). Three strategies were developed and combined for language translation to implement the final version of the language translator module. The summative evaluation, carried out with Deaf from Madrid and Toledo, includes objective measurements from the system and subjective information from questionnaires. The paper also describes the first Spanish-LSE parallel corpus for language processing research focused on specific domains. This corpus includes more than 4000 Spanish sentences translated into LSE. These sentences focused on two restricted domains: the renewal of the identity document and driver’s license. This corpus also contains all sign descriptions in several sign-writing specifications generated with a new version of the eSign Editor. This new version includes a grapheme to phoneme system for Spanish and a SEA-HamNoSys converter.

Bayesian networks, BNs, are suitable for mixed-initiative dialog modeling allowing a more flexible and natural spoken interaction. This solution can be applied to identify the intention of the user considering the concepts extracted from the last utterance and the dialog context. Subsequently, in order to make a correct decision regarding how the dialog should continue, unnecessary, missing, wrong, optional and required concepts have to be detected according to the inferred goals. This information is useful to properly drive the dialog prompting for missing concepts, clarifying for wrong concepts, ignoring unnecessary concepts and retrieving those required and optional. This paper presents a novel BNs approach where a single BN is obtained from N goal-specific BNs through a fusion process. The new fusion BN enables a single concept analysis which is more consistent with the whole dialog context.

In this paper we describe the acoustic emotion recognition system built at the Speech Technology Group of the Universidad Politecnica de Madrid (Spain) to participate in the INTERSPEECH 2009 Emotion Challenge. Our proposal is based on the use of a Dynamic Bayesian Network (DBN) to deal with the temporal modelling of the emotional speech information. The selected features (MFCC, F0, Energy and their variants) are modelled as different streams, and the F0 related ones are integrated under a Multi Space Distribution (MSD) framework, to properly model its dual nature (voiced/unvoiced). Experimental evaluation on the challenge test set, show a 67.06%and 38.24% of unweighted recall for the 2 and 5-classes tasks respectively. In the 2-class case, we achieve similar results compared with the baseline, with a considerable less number of features. In the 5-class case, we achieve a statistically significant 6.5% relative improvement

Nowadays, current commercial and academic platforms for developing spoken dialogue applications lack of acceleration strategies based on using heuristic information from the contents or structure of the backend database in order to speed up the definition of the dialogue flow.In this paper we describe our attempts to take advantage of these information sources using the following strategies: the quick creation of classes and attributes to define the data model structure, the semi-automatic generation and debugging of database access functions, the automatic proposal of the slots that should be preferably requested using mixed-initiative forms or the slots that are better to request one by one using directed forms, and the generation of automatic state proposals to specify the transition network that defines the dialogue flow.Subjective and objective evaluations confirm the advantages of using the proposed strategies to simplify the design, and the high acceptance of the platform and its acceleration strategies.

In the task of speaker diarization for meetings it has been shown in previous work that it is useful to use the time delay of arrival (TDOA) between the different audio channels in the meeting room as an extra source of information in addition to the acoustic features. When combining feature streams, we use a weight to control the relative contributions of the streams. In the past, this weight was determined using development data and the same weight value was applied to all meetings. In this paper we present a method for automatically determining the weight. A metric derived from the Bayesian information criterion (BIC) computed for each feature stream estimates the weight for each meeting on the initial clustering iteration and adapts its value throughout the diarization process. By using this technique we achieve a more robust system and up to 18.2% relative improvement over the method of tuning the weight on development data.

The growing interest in emotional speech synthesis urges effective emotion conversion techniques to be explored. This paper estimates the relevance of three speech components (spectral envelope, residual excitation and prosody) for synthesizing identifiable emotional speech, in order to be able to customize voice conversion techniques to the specific characteristics of each emotion. The analysis has been based on a listening test with a set of synthetic mixed-emotion utterances that draw their speech components from emotional and neutral recordings. Results prove the importance of transforming residual excitation for the identification of emotions that are not fully conveyed through prosodic means (such as cold anger or sadness in our Spanish corpus).

Human-machine interaction in meetings requires the localization and identification of the speakers interacting with the system as well as the recognition of the words spoken. A seminal step toward this goal is the field of rich transcription research, which includes speaker diarization together with the annotation of sentence boundaries and the elimination of speaker disfluencies. The sub-area of speaker diarization attempts to identify the number of participants in a meeting and create a list of speech time intervals for each such participant. In this paper, we analyze the correlation between signals coming from multiple microphones and propose an improved method for carrying out speaker diarization for meetings with multiple distant microphones. The proposed algorithm makes use of acoustic information and information from the delays between signals coming from the different sources. Using this procedure, we were able to achieve state-of-the-art performance in the NIST spring 2006 rich transcription evaluation, improving the Diarization Error Rate (DER) by 15% to 20% relative to previous systems.

Abstract. In this paper we present the ICSI speaker diarization system submitted for the NIST Rich Transcription evaluation (RT06s) [1] con-ducted on the meetings environment. The presented system is based on the RT05s system, which uses agglomerative clustering with a modified Bayesian Information Criterion (BIC) measure to decide which pairs of clusters to merge and to determine when to stop merging clusters. In this year’s system we have eliminated any remaining need for training data, therefore increasing robustness. In our primary system we have intro-duced several improvements from last year. First, we use a new training-free speech/non-speech detection algorithm. Second, we introduce a new algorithm for system initialization. The third improvement is the use of a frame purification algorithm to increase cluster discriminability. Fi-nally, we describe the use of inter-channel delays as features. We explain each of these improvements and show our system’s results on the official evaluation data using hand-aligned references and forced-alignments. We also analyze some of the results and propose improvements. 1

Speaker diarization for recordings made in meetings consists of identifying the number of participants in each meeting and creating a list of speech time intervals for each participant. In recently published work [7] we presented some experiments using only TDOA values (Time Delay Of Arrival for different channels) applied to this task. We demonstrated that information in those values can be used to segment the speakers. In this paper we have developed a method to mix the TDOA values with the acoustic values by calculating a combined log-likelihood between both sets of vectors. Using this method we have been able to reduce the DER by 16.34 % (relative) for the NIST RT05s set (scored without overlap and manually transcribed references) the DER for our devel06s set (scored with overlap and force-aligned references) by 21 % (relative) and the DER for the NIST RT06s (scored with overlap and manually transcribed references) by 15 % (relative). Index terms: Speaker diarization, speaker segmentation, meetings recognition. 1.

In this paper, we present a complete platform for the semiautomatic generation of human-machine dialog systems, that using as input a description of the database of the service, a flow model with the different states of the final application and a guided interaction step by step with the designer’s intervention, generates dialogs to access the service data in different languages and two modalities, speech and web, simultaneously. We describe in detail several strategies that have been followed to reduce the time needed to do the design using the mentioned information. We also address important issues in dialog applications as mixed initiative and overanswering dialogs, confirmation handling and how to provide the user long lists of information.

The EC funded research project GEMINI (Generic Environment for Multilingual Interactive Natural Interfaces) has two main objectives: On the one hand the development and implementation of a platform able to produce user-friendly interactive multilingual and multi-modal dialogue interfaces to databases with a minimum of human effort and on the other hand the demonstration of the platform’s efficiency through the development of two different applications using this platform. The platform consists of different assistants that help the user to semi-automatically generate dialogue applications. Its open and modular architecture simplifies the adaptability of generated applications to different use cases. 1.

This article describes a first version of a system for translating speech into Spanish Sign Language. The system proposed is made up of 4 modules: speech recognizer, semantic analysis, gesture sequence generation and gesture playing. For the speech recognizer and the semantic analysis, we use modules developed by IBM and the University of Colorado respectively. The gesture sequence generation uses the semantic concepts (obtained in the semantic analysis) associating them to several Spanish Sign Language gestures. This association is carried out based on a number of generating rules. For gesture animation, we have developed an animated character and a strategy for reducing the effort in gesture generation. This strategy consists of making the system generate automatically all agent positions necessary for the gesture animation. In this process, the system uses a few main agent positions (2-3 per second) and some interpolation strategies, both issues previously generated by the service developer. 1.

In this paper, we present the work done in language identification for two air traffic control speech recognizers, one for continuous speech and the other one for a command interface. The system is able to distinguish between Spanish and English. We will confirm the advantage of using PPRLM over PRLM. All previous studies show that PPRLM is the technique with the best performance despite of its drawbacks: more processing time and labeled data is needed. No work has been published regarding the optimum weights which should be given to the language models to optimize the performance of the language recognizer. This paper addresses this topic, providing three different approaches for weight selection in the language model score. We will also see that a trigram language model improves performance. The final results are very good even with very short segments of speech. 1.

In this paper we present a whole set of improvements that have been applied to a large vocabulary isolated-word recognition system using continuous models. This system has been used in the EU funded IDAS project (LE4-8315), where an automated interactive telephone-based directory assistance service has been developed.

A number of recent studies have focused on the conceptualized expression of emotions as a three-dimensional space. This paper proposes a new approach to emotion recognition, making use of two of the emotional dimensions and their relationship with different kinds of features. The main idea consists in associating prosodic features, derived from pitch, loudness, and duration, with the activation or arousal dimension, and quality features, i.e. phonation type, articulation manner, voice timbre, with the evaluation and pleasure dimension, in a way that different classification methods can be applied for each specific case. Most important results are achieved for the speaker-independent case and three classes, with a recognition rate close to 80%. 1.

The objective of this paper is the accurate prediction of segmental duration in a Spanish text-to-speech system. There are many parameters that affect duration, but not all of them are always relevant. We present a complete environment in which to decide which parameters are more relevant and the best way to code them. This work is the continuation of Cordoba et al. (1999), where all efforts were dedicated to an unrestricted-domain database for a male voice. In this case, we are considering a female voice in a restricted-domain environment. This restricted-domain offers several advantages to the modeling: the variation in the different patterns is reduced, and so most of the decisions we have made about the parameters are now based in more significant results. So, the conclusions that we present now show clearly which parameters are best. The system is based in a neural network absolutely configurable.

We present the analysis of a Spanish prosody database by estimating the parameters of Fujisaki's (1981) model for F/sub 0/ contours. These parameters are classified attending to linguistic features and they form the analysis database. When synthesizing F/sub 0/ contours we extract the linguistic features from the text and perform a k-nearest neighbour search. Linguistic feature comparison distance is trained using data from the prosody database. To avoid artifacts we perform a rule-base filtering on synthesis parameters. The results of our evaluation test show that the proposed system is significantly better than the previous neural network approach. This evaluation confirms the ability of Fujisaki's model to represent prosody information based on linguistic features.

This paper provides improved confidence assessment for detection of word-level speech recognition errors, out of domain utterances and incorrect concepts in the CU Communicator system. New features from the speech understanding component are proposed for confidence annotation at the utterance and concept levels. We consider a neural network to combine all features in each level. Using the data collected from a live telephony system, it is shown that 53.2% of incorrectly recognized words, 53.2% of out of domain utterances and 50.1% of incorrect concepts are detected at a 5% false rejection rate. In addition, the confidence measures are used to improve the word recognition accuracy. Several hypotheses from different speech recognizers are compiled into a word-graph. The word-graph is searched for the hypothesis with the best confidence. We report a 14.0% relative word error rate reduction after this confidence rescoring.

In the EU funded IDAS project (LE4-8315), demonstrators providing an automated interactive telephone-based directory assistance service have been developed by ten partners from Germany, Greece, Spain and Switzerland [6]. In this paper we will focus in the Spanish demonstrator. In particular, we will describe the following aspects: The general architecture of the system, paying special attention to the speech recognition module. We will present new alternatives for the estimation of continuous HMMs and the agglomerative clustering of context-dependent units.

We present a multi-speaker formant synthesizer based on parameter concatenation. The user can choose among three speakers, two males and one female. The synthesizer stores all the parameters for the basic speaker and linear transformation functions to synthesized the other two. The complete database for one speaker consists of 455 parameterized units (diphones, triphones,...) and the parameters used are pitch, formants and bandwidths and source parameters (four parameters for the LF model, and glottal noise). To get the converted speaker we store a linear transformation function for each spectral stable segment of each unit. Preliminary results show that the quality of the synthesizer is very good and that this system can help us to study and understand the speaker variability problem. 1.

In this paper, we propose a new methodology for designing dialogue managers in telephone-based spoken dialogue systems. This methodology comprises five steps: database analysis, design by intuition, design by observation, simulation and iterative improvement. At each step, several measures to evaluate the designing alternatives are presented. We introduce confidence measures in recognition to define an efficient confirmation strategy in each case. The use of usermodeling techniques adapts the system to the user ability. This methodology is applied for designing a telephone-based system that provides rail travel information for the main Spanish intercity connections, including timetables, simulated fares and reservations. With 30 users completing 4 scenarios, the average duration for a fully automatic call was 204 seconds. The users validated the applicability and usability of the system with a global score of 3.9 (out of 5).

This paper provides improved confidence assessment for detection of word-level speech recognition errors and out-of-domain user requests using language model features. We consider a combined measure of confidence that utilizes the language model back-off sequence, language model score, and phonetic length of recognized words as indicators of speech recognition confidence. The paper investigates the ability of each feature to detect speech recognition errors and out-of-domain utterances as well as two methods for combining the features contextually: a multi-layer perceptron and a statistical decision tree. We illustrate the effectiveness of the algorithm by considering utterances from the ATIS airline information task as either in-domain and out-of-domain for the DARPA Communicator task. Using this hand-labeled data, it is shown that 27.9% of incorrectly recognized words and 36.4% of out-of-domain phrases are detected at a 2.5% false alarm rate.

In the context of large vocabulary speech recognition system, it’s of major interest to classify every utterance as being correctly or incorrectly recognised. In this paper we are presenting a preliminary study on a word-level confidence estimation system based on the output of a neural network. We use a combination of multiple features extracted from the acoustical and lexical decoders of our reference system, those available in the hypothesis stage of a hypothesis-verification very large vocabulary telephone speech recognition system. We will show the system archicture, describe the experiments leading to the selection of the set of parameters to be used by the NN and the final performance, showing promising results as compared with the use of standard log-likelihood ratio techniques for confidence scoring. 1.

In interactive speech applications the expected vocabulary and the expected user utterances change from one dialogue step to the next one. The use of several context dependent language models results in a better system performance than the use of a single model. In this paper we present a new approach combining syntactical and statistical language constraints to a single language model. Recognition results on a database of spelled city names are presented. Furthermore a match against the list of all possible city names is performed.

At ICSLP'98 we presented some preliminary results on automatic preselection list length estimation using parametric and non-parametric techniques, for a flexible large and very large vocabulary, speaker independent, isolated-word hypothesis generation system in a telephone environment, with vocabularies of up to 10000 words. In the baseline system, the preselection module generates a fixed-length list of candidate words, to be given to the verification stage. Our idea is making this length variable, depending on any known-in-advance system parameter, to allow decreasing computational demands. In this paper we present a novel approach to preselection list length estimation. A neural network is used to give an initial estimate of the required length, which is further processed to obtain a final value. The key factor to evaluate different methods is calculating the average preselection list length (effort) while keeping the required error rate. 1 INTRODUCTION Computational demands are...

This paper explores the fourth order cumulants (FOC) of the LPC residual of speech signals and presents a new algorithm for Voice Activity detection (VAD) based on the newly established FOC properties. Analytical expressions for the horizontal slice of the 4th cumulant as well as the kurtosis of voiced speech are derived based on a reported sinusoidal model [4]. The derivations demonstrate that the kurtosis of voiced speech is distinct from that of Gaussian noise and can be used to aid in detecting voicing. The proposed VAD combines FOC metrics with SNR measures to classify speech and noise frames. Its performance is compared to the ITU-T G.729B VAD [1] in various noise conditions, and quantified using the probability of correct and false classifications. The results show the proposed VAD has overall comparable performance to the G.729B: Its probability of false classification is lower in low SNR and Gaussian-like noise, but higher in speech-like noises.

In a limited domain task (e.g. airline reservation, database retrieval, etc) many robust understanding systems, designed for both speech and text input, have been implemented [1][3][5][6][10] based on the Stochastic Conceptual FiniteState paradigm (Semantic Network) or CHRONUS paradigm [11] (Conceptual Hidden Representation of Natural Unconstrained Speech), which establishes relations between conceptual entities through a probabilistic graph-like structure. The use of this kind of grammar to model semantic information presents limitations, which have been analysed during the implementation of a flexible architecture for a robust information retrieval system, based on the same paradigm [1]. We have tried to solve some of them by integrating a set of conceptual probabilistic and nonprobabilistic grammars, which allow certain complexity in the functionality of the application, such as applying non-SQL functions to the results of SQL queries in order to retrieve information not explicitly included in the database, translating certain natural spoken sentences (that would produce difficult embedded queries and therefore more natural queries) without so many restrictions in the relative position of inter-concept relationship.

At ICSLP’96 we presented a flexible, large vocabulary, speaker independent, isolated-word preselection system in a telephone environment, using a two stage, bottom-up strategy [6]. We achieved reasonable performance in large and very large vocabulary tasks, ranging from 1200 to 10000 words. In this paper, we describe recent studies we have carried out on the system, aimed at two directions: handling of non speech sounds in the speech signal (we consider lips, respiration and click noises); and making the preselection lists dynamic in length, to reduce computational load, in the average. In the first case, we want to model non speech sounds, as these effects are crucial in real-life situations, leading to wrong endpointing and increasing error rates. In the second, we are interested in integrating any available system parameter to calculate the preselection list length to use, having applied both parametric and non parametric methods. 1.

Modern Speech synthesisers have achieved a high degree of intelligibility, but can not be regarded as natural-sounding devices. In order to decrease the monotony of synthetic speech, the implementation of emotional effects is now being progressively considered. This paper presents a through study of emotional speech in Spanish, and its application to TTS, presenting a prototype system that simulates emotional speech using a commercial synthesiser. The design and recording of a Spanish database will be described and also the analysis of the emotional prosody (by fitting the data to a formal model). Using this collected data, a rule-based simulation of three primary emotions was implemented in the Text-to-Speech system. Finally, the assessment of the synthetic voice through perception experiments will classify the system as capable of producing quality voice with recognisable emotional effects.

We present a new language model that includes some of the most promising techniques for overcoming linguistic inadequacy, including POS tagging [3] and refining [4], hierarchical, locally conditioned grammars [5], parallel modelling of acoustic and linguistic domains [6] – and some of our own: language modelling as language parsing, and a better integration of the whole process with the acoustic model resulting in a richer educt from the language modelling process. We are building this model for a translation into Spanish of the DARPA RM task, maintaining the same 1k words vocabulary and some 1000 sentences.

In this paper, we present a flexible architecture to implement a robust information retrieval system based on domain and linguistic modelling by means of a set of conceptual probabilistic and non-probabilistic grammars. It allows certain complexity in the functionality of the application, such as applying non-SQL functions to the results of SQL queries in order to retrieve information not explicitly included in the database, or translating certain natural spoken sentences that would produce difficult embedded queries.

In this paper we present a modification of the usual decimation-interpolation steps for resampling of speech signals which is especially adapted to arbitrary modification of fundamental frequency and duration of speech segments. The modification is intended to overcome the time and frequency domain limitation that such a resampling scheme imposes so it can be used in a speech synthesis system. The performance of this resampling method for prosody modification is better than the equivalent PSOLA (Pitch-Synchronous Overlap-Add) method when working at a sampling frequency of 8 to 10 kilohertz so the source spectrum of the voiced allophones can be said to be completely harmonical. An optimization of the proposed algorithm that allows a real time implementation is also discussed.

We present the development and characteristics of a basic ASR system for isolated digits in Spanish, used over the telephone line. Initially we will introduce our first idea, a basic discrete system, and then we will see the improvements we made to increase the recognition rate at a low CPU cost (always considering its practical implementation as a real time system). The most remarkable advances were obtained with: 1) Semicontinuous modelling. It is a more precise modelling, although more time consuming. 2) End-pointing with a Neural network. 3) One pass decoding with noise models. The intention of both 2 and 3 is to alleviate the effects of a wrong end-pointing. 4) Parametrization using perceptual filters in frequency and filtering in the time domain (RASTA-PLP). We wanted to decrease the effect of telephonic noise in our system. 1. EXPERIMENTAL SET-UP Our database consists of 1320 repetitions of the ten digits and a break command in Spanish (that makes an average of 120 reps./...

The final quality of a concatenation synthesis system is directly related to the continuity of the spectrum at the concatenation point. Due to the subjective auditory masking, if we minimize the spectral distortion in the formant frequencies, the quality will increase significantly. We present, along with results concerning pitch marking, an algorithm capable of modifying the LPC envelope in a flexible way which is the heart of a spectral smoothing module for a diphone-based linear prediction pitch-synchronous overlap-add (LP-PSOLA) concatenation system.>

The first prototype of a low cost dictation machine for Spanish is described (DIVO). The main characteristics of our recognition approach are: bottomup, hypothesis-verification strategy; large vocabulary, speaker dependent, isolated word recognition. Its modular structure is the cue for quick development and testing of different implementation alternatives. Two of them are presented: one is based in Static phoneme Modeling (SM) and the other uses Discrete Hidden Markov Modeling (DHMM). The system runs on a standard PC compatible (286 or higher) equipped with a DSP board and is fully voice controlled. This first version of the system can address multiple vocabulary sets of up to 2000 words each, with immediate response and reasonable performance. Modules for increasing vocabulary and performance are being developed. I.

We are building a large vocabulary, isolated word preselection system according to a bottom-up design strategy. It will be used in the development of a dictation machine for Spanish and it is composed of three main modules: feature extraction, phonetic string build up and lexical access. In the second one, we are considering three different technological approaches based on static modeling (SM), Hidden Markov Models (HMM) and Neural Networks (NN). This paper will compare these three alternatives in terms of recognition performance, training complexity and computational load, and will conclude with the results of the comparison in order to adopt the most suitable approach depending on the task. I. INTRODUCTION The study we are presenting was done to help the decision process of adopting a certain technology for future developments in very large vocabulary speech recognition systems based in the hypothesisverification paradigm. The stage analyzed here constitutes the first step (hypoth...

Two autonomous low cost devices which extract in real time some parameters of the voice, and perform a visual representation on a screen, are here described. Even though they can be useful in any voice related activity, they are in use at the moment in speech teaching to deaf people with good results. The first one, called SAS, displays a cross-section of the vocal tract and is based on the method proposed by Wakita (1); while the second one, named ISO-TON, makes a representation along the time of whether intensity, voiced/fricative condition of the voice, and fundamental tone.

The objective of the paper is to describe an algorithm for finding robust speech boundaries which can be used as a first step in an acoustic-phonetic-based speech recognition scheme. This algorithm uses information about the speech processing made by the peripheral auditory system. The algorithm has been tested with success with a set of monosyllabic CVC words in English including all the combinations between eleven different consonants and three different vowels. It can find all the vowel onsets, vowel offsets and stop releases for these words.

A complete visual aid for training hearing handicapped based on real time representation of linear prediction estimated vocal tract shapes, has been designed and evaluated with deaf children. Some problems related to this project are discussed in this paper like the areas normalization and the error criterion. Preliminary results of the evaluation are also reported.