Patent
Method and configuration for determining a descriptive feature of a speech signal
TLDR
In this article, a first speech model is trained with a first time pattern and a second speech model with a second time pattern, and the second model is initialized with the first model.Abstract:
A method and also a configuration for determining a descriptive feature of a speech signal, in which a first speech model is trained with a first time pattern and a second speech model is trained with a second time pattern. The second speech model is initialized with the first speech model.read more
Citations
More filters
Patent
Intelligent Automated Assistant
Thomas R. Gruber,Adam Cheyer,Dag Kittlaus,Didier Rene Guzzoni,Christopher Dean Brigham,Richard Donald Giuli,Marcello Bastea-Forte,Harry J. Saddler +7 more
TL;DR: In this article, an intelligent automated assistant system engages with the user in an integrated, conversational manner using natural language dialog, and invokes external services when appropriate to obtain information or perform various actions.
Patent
Using context information to facilitate processing of commands in a virtual assistant
TL;DR: In this article, a virtual assistant uses context information to supplement natural language or gestural input from a user, which helps to clarify the user's intent and reduce the number of candidate interpretations of user's input, and reduces the need for the user to provide excessive clarification input.
Patent
Method and apparatus for building an intelligent automated assistant
Adam Cheyer,Didier Rene Guzzoni +1 more
TL;DR: In this paper, a method for building an automated assistant includes interfacing a service-oriented architecture that includes a plurality of remote services to an active ontology, where the active ontologies includes at least one active processing element that models a domain.
Patent
Automatically adapting user interfaces for hands-free interaction
TL;DR: In this article, the authors present a method for automatically determining whether a digital assistant application has been separately invoked by a user without regard to whether a user has separately invoked the application.
Patent
Voice trigger for a digital assistant
TL;DR: In this paper, a method for operating a voice trigger is presented, which includes determining whether at least a portion of the sound input corresponds to a predetermined type of sound, such as a human voice.
References
More filters
Patent
Dimensionality reduction for speaker normalization and speaker and environment adaptation using eigenvoice techniques
TL;DR: In this article, a set of speaker dependent models or adapted models is trained upon a comparatively large number of training speakers, one model per speaker, and model parameters are extracted in a predefined order to construct a sets of supervectors, one per speaker.
PatentDOI
Speech recognition system
TL;DR: In this article, an improved speech recognition system with at least two speech recognition engines which may or may not be identical is presented, each of which provides a recognized-text output signal, which are provided to a text comparator.
Multi-Stream Speech Recognition
TL;DR: This paper introduces the basic framework of a statistical structure that can accommodate multiple (asynchronous) observation streams (possibly exhibiting different frame rates) and will then be applied to the particular case of multi-band speech recognition and will be shown to yield significantly better noise robustness.
Patent
Multi-resolution system and method for speaker verification
TL;DR: In this paper, a method for generating a speaker-dependant model of an utterance that has at least one occurrence is presented, which includes generating an initial model, having a first resolution, that encodes each of the occurrences of the utterance; and generating at least an additional speaker-specific model having a different resolution from that of the initial model.
Patent
Method of independently creating and using a garbage model for improved rejection in a limited-training speaker-dependent speech recognition system
TL;DR: Agarwal et al. as discussed by the authors proposed a speaker-dependent (SD) speech recognition system with very little training data, and also within hardware constraints such as limited memory and processing resources.