Fujitsu Corporation

Yusuke Hamada

Takahiro Shikibu

Satoru Sankoda

Osamu Moriyama

Hiroto Takegawa

Keisuke Asakura

Misato Hishi

Kousuke Iemura

A speech recognition method, performed by a computer, with an improved recognition accuracy is disclosed. The method includes: performing speech recognition of an input speech to acquire a plurality of recognition candidates through a plurality of speech recognition processes different from each other for a section having a reliability lower than a predetermined value; verifying similarities between each of the acquired plurality of recognition candidates and meta-information corresponding to the input speech; and determining, based on the verified similarities, a recognition result of the low-reliability section from among the acquired plurality of recognition candidates.

Speech recognition apparatus, speech recognition program, and speech recognition method

A speech recognition method causes a computer to execute a speech recognition process. The process includes: extracting features of noise or unwanted sound included in speech data to be recognized; performing, for the speech data, a suppression process for each of the extracted features of noise or unwanted sound among suppression processes for respective features of noise or unwanted sound; and performing speech recognition for the speech data after the suppression process is performed.

Speech recognition apparatus, speech recognition method, and a recording medium performing a suppression process for categories of noise

A character recognition method for a moving image includes extracting a region corresponding to a character string included in each frame of a moving image to be recognized. The method reads a character string from the extracted region and corrects the character string read from each frame, based on appearance rule information that specifies an appearance rule of a character string corresponding to the order of the frames, such that the appearance order of the read character string conforms to the appearance rule.

Character recognition apparatus, character recognition program, and character recognition method

An apparatus generates first context data representing a context of correction target text based on the correction target text, and corrects an error in the correction target text by inputting a character string of the correction target text, the generated first context data, and meta-information corresponding to the correction target text to a neural network that has been trained to correct an error in the correction target text by inputting a character string of text corresponding to training data, second context data representing a context of the text, and meta-information of the text.

Accurate correction of errors in text data based on learning via a neural network

A semiconductor device, includes: a first semiconductor chip including: a first substrate; a first via; a first rear surface-side pad connected to the first via; a first wiring layer; a first front surface-side pad formed on the first wiring layer; and an input circuit formed in the first substrate, an input signal wire connecting the first via, the first front surface-side pad, and an input terminal of the input circuit; and a second semiconductor chip including: a second substrate; a second wiring layer; a second front surface-side pad; and an output circuit formed in the second substrate, an output signal wire connecting the second front surface-side pad to an output terminal of the output circuit. The second semiconductor chip is stacked on a rear surface side of the first semiconductor chip, and the first rear surface-side pad and the second front surface-side pad are connected.

Semiconductor device having a plurality of chips being stacked

A semiconductor device includes semiconductor chips stacked each other. Each of the semiconductor chips converts second reception data received by second reception terminals arranged in point symmetry on the first face by a conversion method to convert first reception data received by first reception terminals arranged in point symmetry on the first face into a reference data; and generates an identification information of the each semiconductor chip based upon the converted second reception data; and outputs the bit sequence obtained by converting the generated identification information by means of the inverse conversion method of the conversion method.

Stacked semiconductor device and control method for the same

A semiconductor device includes: a first semiconductor chip including a first terminal at a first face side, a first load whose one end is connected to the first terminal, another end of the first load being to be connected to a power source potential, a second terminal at a second face side, a second load whose one end is connected to the second terminal, another end of the second load being to be connected to a ground potential, a first detection circuit that detects generation of potential difference at the first load, and a second detection circuit that detects generation of potential difference at the second load; and a second semiconductor chip including a connection terminal disposed at a face facing the first semiconductor chip; wherein the power source potential or the ground potential is to be connected through the connection terminal to the first or second terminal.

Growing community of inventors

Ota, Japan

Yusuke Hamada