Abstract: A method for compensating inter-session variability for automatic extraction of information from an input voice signal representing an utterance of a speaker, includes: processing the input voice signal to provide feature vectors each formed by acoustic features extracted from the input voice signal at a time frame; computing an intersession variability compensation feature vector; and computing compensated feature vectors based on the extracted feature vectors and the intersession variability compensation feature vector.

Abstract: A method for coding data, includes: grouping data into frames; classifying the frames into classes; for each class, transforming the frames belonging to the class into filter parameter vectors, which are extracted from the frames by applying a first mathematical transformation; for each class, computing a filter codebook based on the filter parameter vectors belonging to the class; segmenting each frame into subframes; for each class, transforming the subframes belonging to the class into source parameter vectors, which are extracted from the subframes by applying a second mathematical transformation based on the filter codebook computed for the corresponding class; for each class, computing a source codebook based on the source parameter vectors belonging to the class; and coding the data based on the computed filter and source codebooks.

Abstract: A text-to-speech system adapted to operate on text in a first language including sections in a second language, includes a grapheme/phoneme transcriptor for converting the sections in the second language into phonemes of the second language; a mapping module configured for mapping at least part of the phonemes of the second language onto sets of phonemes of the first language; and a speech-synthesis module adapted to be fed with a resulting stream of phonemes including the sets of phonemes of the first language resulting from mapping and the stream of phonemes of the first language representative of the text, and to generate a speech signal from the resulting stream of phonemes.

Abstract: A text-to-speech system adapted to operate on text in a first language including sections in a second language, includes a grapheme/phoneme transcriptor for converting the sections in the second language into phonemes of the second language; a mapping module configured for mapping at least part of the phonemes of the second language onto sets of phonemes of the first language; and a speech-synthesis module adapted to be fed with a resulting stream of phonemes including the sets of phonemes of the first language resulting from mapping and the stream of phonemes of the first language representative of the text, and to generate a speech signal from the resulting stream of phonemes.

Abstract: A system for access to multimedia structures has telephone sets capable of connecting to a telephone network, a storage device capable of storing a plurality of multimedia structures representing messages and/or data and/or commands, and a network access server that can be associated with the telephone sets and is capable of selectively instantiating the multimedia structures via an interconnection network. There is also a voice-recognition and speech-synthesis system that can be associated with the network access server and that comprises modules for reading files in XML format and for processing the files so as to obtain files in a format that can be synthesized by a speech synthesizer.

Abstract: A method of adapting a neural network of an automatic speech recognition device, includes the steps of: providing a neural network including an input stage, an intermediate stage and an output stage, the output stage outputting phoneme probabilities; providing a linear stage in the neural network; and training the linear stage by means of an adaptation set; wherein the step of providing the linear stage includes the step of providing the linear stage after the intermediate stage.

Abstract: A text-to-speech system adapted to operate on text in a first language including sections in a second language, includes a grapheme/phoneme transcriptor for converting the sections in the second language into phonemes of the second language; a mapping module configured for mapping at least part of the phonemes of the second language onto sets of phonemes of the first language; and a speech-synthesis module adapted to be fed with a resulting stream of phonemes including the sets of phonemes of the first language resulting from mapping and the stream of phonemes of the first language representative of the text, and to generate a speech signal from the resulting stream of phonemes.

Abstract: Grapheme-to-phoneme alignment quality is improved by introducing a first preliminary alignment step, followed by an enlargement step of the grapheme-set and phoneme-set, and a second alignment step based on the previously enlarged grapheme /phoneme sets. During the enlargement step, grapheme clusters and phoneme clusters are generated that become members of a new grapheme and phoneme set. The new elements are chosen using statistical information calculated using the results of the first alignment step. The enlarged sets are the new grapheme and phoneme alphabet used for the second alignment step. The lexicon is rewritten using this new alphabet before starting with the second alignment step that produces the final result.

Abstract: A system for access to multimedia structures has telephone sets capable of connecting to a telephone network, a storage device capable of storing a plurality of multimedia structures representing messages and/or data and/or commands, and a network access server that can be associated with the telephone sets and is capable of selectively instantiating the multimedia structures via an interconnection network. There is also a voice-recognition and speech-synthesis system that can be associated with the network access server and that comprises modules for reading files in XML format and for processing the files so as to obtain files in a format that can be synthesized by a speech synthesizer.

Abstract: An automatic speech recognition method for identifying words from an input speech signal includes providing at least one hypothesis recognition based on the input speech signal, the hypothesis recognition being an individual hypothesis word or a sequence of individual hypothesis words, and computing a confidence measure for the hypothesis recognition, based on the input speech signal, wherein computing a confidence measure includes computing differential contributions to the confidence measure, each as a difference between a constrained acoustic score and an unconstrained acoustic score, weighting each differential contribution by applying thereto a cumulative distribution function of the differential contribution, so as to make the distributions of the confidence measures homogeneous in terms of rejection capability, as the language, vocabulary and grammar vary, and computing the confidence measure by averaging the weighted differential contributions.

Abstract: A system for access to multimedia structures has telephone sets capable of connecting to a telephone network, a storage device capable of storing a plurality of multimedia structures representing messages and/or data and/or commands, and a network access server that can be associated with the telephone sets and is capable of selectively instantiating the multimedia structures via an interconnection network. There is also a voice-recognition and speech-synthesis system that can be associated with the network access server and that comprises modules for reading files in XML format and for processing the files so as to obtain files in a format that can be synthesized by a speech synthesizer.

Abstract: A neural network in a speech-recognition system has computing units organized in levels including at least one hidden level and one output level. The computing units of the hidden level are connected to the computing units of the output level via weighted connections, and the computing units of the output level correspond to acoustic-phonetic units of the general vocabulary. This network executes the following steps: determining a subset of acoustic-phonetic units necessary for recognizing all the words contained in the general vocabulary subset; eliminating from the neural network all the weighted connections afferent to computing units of the output level that correspond to acoustic-phonetic units not contained in the previously determined subset of acoustic-phonetic units, thus obtaining a compacted neural network optimized for recognition of the words contained in the general vocabulary subset; and executing, at each moment in time, only the compacted neural network.

Abstract: A method of optimizing the execution of a neural network in a speech recognition system provides for conditionally skipping a variable number of frames, depending on a distance computed between output probabilities, or likelihoods, of a neural network. The distance is initially evaluated between two frames at times 1 and 1+k, where k is a predetermined maximum distance between frames, and if such distance is sufficiently small, the frames between times 1 and 1+k are calculated by interpolation, avoiding further executions of the neural network. If, on the contrary, such distance is not small enough, it means that the outputs of the network are changing quickly, and it is not possible to skip too many frames. In that case, the method attempts to skip remaining frames, calculating and evaluating a new distance.

Abstract: Syntagms of a text including individual elements written without separators are segmented into chunks having strings including at least one individual element, such as an ideogram of the Mandarin Chinese language. A lexicon is defined including a set of strings, each string having at least one of the individual elements. The syntagm, being segmented, is orderly searched on an element-by-element basis by searching within the lexicon strings corresponding to any of the chunks. In the case of a positive search result, the corresponding chunk located is stored with an associated cost. A check is made as to whether the chunk located was already present in the lexicon. If the chunk located was already present, the cost associated therewith is reduced. A plurality of candidate segmentation sequences are thus generated, each corresponding to a respective segmentation pattern having associated a corresponding accrued cost.

Abstract: Method for managing mixed-initiative human-machine dialogues based on speech interaction exploiting the separation between general dialogue knowledge, such as communicative acts, which can be used in multiple application domains, and particular linguistic knowledge, which are domain-specific parameters, to process the dialogue as a sequence of changes of status. Each status consist in a set of features linked both to the processed parameters and to the linguistic and pragmatic context, and describes a certain instant of the communicative situation between the user and the system so to discriminate it from other situations that are also only slightly different. The method employs three components.

Abstract: A method for automatically providing a hypothesis of a linguistic formulation that is uttered by users of a voice service based on an automatic speech recognition system and that is outside a recognition domain of the automatic speech recognition system. The method includes providing a constrained and an unconstrained speech recognition from an input speech signal, identifying a part of the constrained speech recognition outside the recognition domain, identifying a part of the unconstrained speech recognition corresponding to the identified part of the constrained speech recognition, and providing the linguistic formulation hypothesis based on the identified part of the unconstrained speech recognition.

Abstract: Disclosed herein is a noise reduction method for automatic speech recognitionl.

Abstract: A method for extracting sampled voice signal features for an automatic voice recognition system essentially comprises the following steps: