Samsung Electronics Co., Ltd.

Jeongmi Cho

Jeongsu Kim

Jaewon Lee

Yongbeom Lee

Kwangil Hwang

Ki-hyun Choo

Inho Kang

Jihye Chung

A multi-layered speech recognition apparatus and method, the apparatus includes a client checking whether the client recognizes the speech using a characteristic of speech to be recognized and recognizing the speech or transmitting the characteristic of the speech according to a checked result; and first through N-th servers, wherein the first server checks whether the first server recognizes the speech using the characteristic of the speech transmitted from the client, and recognizes the speech or transmits the characteristic according to a checked result, and wherein an n-th (2&#x2266;n&#x2266;N) server checks whether the n-th server recognizes the speech using the characteristic of the speech transmitted from an (n&#x2212;1)-th server, and recognizes the speech or transmits the characteristic according to a checked result.

Multi-layered speech recognition apparatus and method

A method, medium, and apparatus for generating a record sentence to establish a speech corpus, including generating a synthesized sentence of speech and synthesis information related to speech synthesis by performing speech synthesis for a predetermined sentence of text, selecting an unseen sentence including an unseen unit according to the synthesis information, generating a weight indicating a recording priority of the unseen unit included in the selected unseen sentence, and generating a record sentence by combining the unseen unit with the speech synthesis information according to the generated weight.

Apparatus, medium, and method for generating record sentence for corpus and apparatus, medium, and method for building corpus using the same

A method and apparatus for updating a dictionary, by which documents of interest of a user are extracted through a network and a dictionary is updated with individual names and coined words extracted from the web documents, are provided. The dictionary updating method includes: extracting a document of interest of a user in each predetermined extraction period from a predetermined server connected to a dictionary updating system through a network, and extracting candidate unregistered words existing in the extracted document according to predetermined unregistered word extraction rules; based on information on appearance frequencies of the candidate unregistered words retrieved from the document, extracting unregistered words among the candidate unregistered words and extracting candidate semantic classes of the unregistered word; with usage examples of the unregistered word obtained through a searching unit, according to a predetermined unregistered word verification method, verifying the unregistered word and determining the semantic class of the verified unregistered word; and updating the dictionary updating system with the verified unregistered word and the semantic class of the verified unregistered word.

Periodically extracting and evaluating frequency of occurrence data of unregistered terms in a document for updating a dictionary

Growing community of inventors

Suwon-si, South Korea

Jeongmi Cho