Pohang University of Science and Technology Foundation

Geun Bae Lee

Dong Hyeon Lee

Yong Kim

Jun Hwi Choi

Seong Han Ryu

Sang Jun Koo

In Jae Lee

Hong Suck Seo

Seok Hwan Kim

Kyung Duk Kim

A method for automatic multilingual speech recognition may comprise: recognizing input audio data by a speech recognizer; classifying the audio data by a speech language classifier; activating, by an output layer selector coupled to the speech recognizer, any one projection output layer of a plurality projection output layers respectively connected to the speech recognizer according to language classification information received from the speech language classifier, an output unit of the activated projection output layer being configured as several bytes; and recombining outputs output in a unit of the several bytes by the activated projection output layer, and outputting the recombined output as an automatic speech recognition result for the audio data.

Method and apparatus for multilingual speech recognition based on artificial intelligence models

Disclosed are a method for processing a dialog based on processing instructing expression in a multi-modal environment and an apparatus therefor. The method for processing a dialog in an information processing device capable of processing digital signals includes the steps of: extracting an instructing expression from an inputted sentence; generating an intermediate instructing expression representing the modifying relations between the words constituting the extracted instructing expression; and searching the object corresponding with the intermediate instructing expression in a predetermined object search range. Thus, a terminal can be effectively and conveniently used without separately clarifying various instructing expressions representing things or objects with the terminal.

Method for processing dialogue based on processing instructing expression and apparatus therefor

The present invention is a method for recognizing a statistical voice language using a statistical technique without using a manually tagged corpus. The method comprises: a dialog act clustering step of clustering speech utterances of sentences based on similar dialog acts; a named entity clustering step of extracting a named entity candidate group from the result of the dialog act clustering step and clustering named entities based on the neighboring contextual information of the extracted named entity candidate group; and a main act clustering step of clustering main acts for each region based on the clustered dialog acts and named entity.

Method for recognizing statistical voice language

Disclosed are a speech recognition error correction method and an apparatus thereof. The speech recognition error correction method includes determining a likelihood that a speech recognition result is erroneous, and if the likelihood that the speech recognition result is erroneous is higher than a predetermined standard, generating a parallel corpus according to whether the speech recognition result matches the correct answer corpus, generating a speech recognition model based on the parallel corpus, and correcting an erroneous speech recognition result based on the speech recognition model and the language model. Accordingly, speech recognition errors are corrected.

Method and apparatus for correcting speech recognition error

According to the present invention, a method for searching for information using the Web and a method for voice conversation using same involve generating a basic word vector for at least one of a provided user query and a language analysis result; using the basic word vector, searching a vector space database for a vector space corresponding to the basic word vector; searching, if a similarity between the basic word vector and the found vector space is lower than a preset reference, a vector space database for a vector space corresponding to an expanded word vector using the expanded word vector generated from the result of the Web search performed using at least one of the user query and the language analysis result; and searching for knowledge information on the basis of the vector space found in the basic searching step or the expanded searching step. Thus, improved research results may be provided in response to the user query.

Growing community of inventors

Pohang-si, South Korea

Geun Bae Lee