Matsushita Electric Industrial Co., Ltd.

Fujitsu Corporation

Tatsuya Kimura

Taisuke Watanabe

Tsuyoshi Inoue

Mitsuru Endo

Akira Ishida

Tomohiro Konuma

Katsuyuki Niyada

Takeo Hayashi

Masakatsu Hoshimi

Kenichi Myokan

Natsuki Saito

Shoji Hiraoka

A method and apparatus for speech recognition of the present application has a process to collate, with an input utterance, an acoustic model corresponding to a hypothesis to be expressed by the connection of utterance segments, such as phonemes or syllables, and developed according to a length of an input utterance by an inter-word connection rule thereby obtaining a recognition score. Within a word of the hypothesis, the similar hypotheses high in utterance score within a predetermined threshold from the maximum value of the score are all held to a word end irrespectively of the number of hypotheses. Meanwhile, at a word end of the hypotheses, the hypotheses are narrowed to a predetermined number of upper ranking in the order of higher score.

Method and apparatus for speech recognition

A disk unit for reading information from or writing information to a disk by means of a head supported by an actuator includes a shroud having a face perpendicular to a surface of the disk and opposing a peripheral edge of the disk, and a spoiler having a given height in a direction perpendicular to the surface of the disk and extending above the surface of the disk from the peripheral edge generally toward a center of the disk.

Disk unit with increased head-positioning accuracy

A method of speech recognition includes the steps of analyzing input speech every frame and deriving feature parameters from the input speech, generating an input vector from the feature parameters of a plurality of frames, and periodically calculating partial distances between the input vector and partial standard patterns while shifting the frame one by one. Standard patterns correspond to recognition-object words respectively, and each of the standard patterns is composed of the partial standard patterns which represent parts of the corresponding recognition-object word respectively. The partial distances are accumulated into distances between the input speech and the standard patterns. The distances correspond to the recognition-object words respectively. The distances are compared with each other, and a minimum distance of the distances is selected when the input speech ends. One of the recognition-object words which corresponds to the minimum distance is decided to be a recognition result.

Method of speech recognition

A method of speech recognition includes the steps of predetermining a series of parameters a.sub.1, a.sub.2, . . . , a.sub.J representative of standard patterns of speeches of preset words, where the letter J denotes a predetermined natural number; deriving parameters x.sub.i representative of data of respective frames of an input signal, the adscript i denotes a frame number; calculating similarities d.sub.j.sup.(i) between the parameters a.sub.j and the parameters x.sub.i, where j=1, 2, . . . , J; calculating parameters R.sub.j.sup.(i) for the respective preset words by referring to the following recurrence formulas:

Power information from an input signal is used to detect the presence of speech. A reference point is established by the power information as the moment of detection of the start of speech. The end of a processing period for speech recognition is determined by a power information occurring after end of speech, so that feature parameters are extracted from the signal during the processing period having start and end points determined by the power information. Standard speech patterns of particular preset words are determined and similarities between the extracted feature parameters and the standard patterns are calculated and mutually compared. The selected preset word corresponds to a maximum of the similarities obtained during the processing period selected in accordance with the power information. The selected word is then outputted as the recognition result. Selection of the process time in accordance with the power information permits exclusion of some similarities in the speech recognition process, thereby preventing erroneous recognition which could occur in a long word having a portion similar to a short word.

Power information from an input signal is used to detect the presence of speech. A reference point is established as the moment of detection of the speech. During a period between the reference point and a subsequent point distant from the reference point by a predetermined range, the input signal is linearly changed to a corresponding signal having a predetermined period. Feature parameters are extracted from the signal with the predetermined period. The feature parameters are replaced by preset noise parameters in a portion having no speech component therein. Standard speech patterns of particular preset words are determined and similarities between the extracted feature parameters containing the noise parameters and the standard patterns are calculated and mutually compared. The foregoing steps are performed while the separations are varied within the predetermined range. Similar steps are performed as the reference point is shifted by a unit period, and similarities are calculated and mutually compared. Speech duration is detected by use of movement of the power information. A process end time is determined by use of the speech duration time and a time dependent variation in the similarities. The selected preset word corresponds to a maximum of the similarities obtained when the reference point reaches the process end time. The selected word is then outputted as the recognition result.

Growing community of inventors

Kawasaki, Japan

Tatsuya Kimura