A system and method of the present invention for determining and maintaining dialog focus in a conversational speech system includes presenting a command associated with an application to a dialog manager The application associated with the command is unknown to the dialog manager at the time it is made The dialog manager determines a current context of the command by reviewing a multi-modal history of events At least one method is determined responsive to the command based on the current context The at least one method is executed responsive to the command associated with the application

Method for determining and maintaining dialog focus in a conversational speech system

An approach to multi-lingual speech recognition that permits different words in an utterance to be from different languages. Words from different languages are represented using different sets of sub-word units that are each associate with the corresponding language. Despite the use of different sets of sub-word units, the approach enables use of cross-word context at boundaries between words from different languages (cross-language context) to select appropriate variants of the subword units to match the context.

Multi-lingual speech recognition with cross-language context modeling

A method for providing a personality for a conversational system includes providing a plurality of attributes for determining a behavior of the conversational system. When a command is presented to the conversational system for execution, the command is responded to by employing the plurality of attributes such that the user experiences an interface with human characteristics.

Personality generator for conversational systems

In accordance with the invention, a method and system for accessing a dialog system employing a plurality of different clients, includes providing a first client device for accessing a conversational system and presenting a command to the conversational system by converting the command to a form understandable to the conversational system. The command is interpreted by employing a mediator, a dialog manager and a multi-modal history to determine the intent of the command based on a context of the command. A second client device is determined based on a predetermined device preference stored in the conversational system. An application is abstracted to perform the command, and the results of the performance of the command are set to the second client device.

Method and system for multi-client access to a dialog system

We introduce polyglot language models, recurrent neural network models trained to predict symbol sequences in many different languages using shared representations of symbols and conditioning on typological information about the language to be predicted. We apply these to the problem of modeling phone sequences---a domain in which universal symbol inventories and cross-linguistically shared feature representations are a natural fit. Intrinsic evaluation on held-out perplexity, qualitative analysis of the learned representations, and extrinsic evaluation in two downstream applications that make use of phonetic features show (i) that polyglot models better generalize to held-out data than comparable monolingual models and (ii) that polyglot phonetic feature representations are of higher quality than those learned monolingually.

Polyglot Neural Language Models: A Case Study in Cross-Lingual Phonetic Representation Learning

In this paper we describe our initial e orts in building a natural language understanding (NLU) system across multiple languages. The system allows users to switch languages seamlessly in a single session without requiring any switch in the speech recognition system. Context dependence is maintained across sentences, even when the user changes languages. Towards this end we have begun building a universal speech recognizer for English and French languages. We experiment with a universal phonology for both French and English with a novel mechanism to handle language dependent variations. Our best results so far show about 5% relative performance degradation for English relative to a unilingual English system and a 9% relative degradation in French relative to a unilingual French system. The NLU system uses the same statistical understanding algorithms for each language, making system development, maintenance, and portability vastly superior to systems built customly for each language.

Jerome Gros

Papers

Towards speech understanding across multiple languages.