中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

14 records (took 0.021 seconds)

    METHOD FOR VOICE RECOGNITION DEVICE
    1.
    发明专利
    METHOD FOR VOICE RECOGNITION DEVICE 失效

    公开(公告)号：JP2000242293A

    公开(公告)日：2000-09-08

    申请号：JP2000036105

    申请日：2000-02-15

    Applicant: MOTOROLA INC

    Inventor： MEUNIER JEFFREY ARTHUR ,  POPPERT DANIEL CHARLES

    IPC: G10L15/14 ,  G10L15/12

    Abstract: PROBLEM TO BE SOLVED: To make storable trace-back information needed for HMM tracing by using a smaller memory by dividing the signal representation of a pronounced voice into frames, allocating the respective frames to respective states by using adjusting algorithm, and discriminating state transition to respective states. SOLUTION: A sound signal received from a microphone 114 is converted by an analog-digital converter 202 into a digital signal. A speech call processor inplements a feature extracting function 204 for the processed digital signal representation of the analog signal output of the microphone 114 to generate one group of feature vectors representing the pronounced voice of the user. And, one feature vector is present for every frame. The processor performs speech recognition 206 or training 207 by making use of the feature vectors. In training mode, the feature vector of the pronounced voice is used to generate a template in HMM format to be stored in a memory 208.

    FEATURE EXTRACTING METHOD FOR SPEECH RECOGNITION SYSTEM
    2.
    发明专利
    FEATURE EXTRACTING METHOD FOR SPEECH RECOGNITION SYSTEM 失效

    公开(公告)号：JP2000250576A

    公开(公告)日：2000-09-14

    申请号：JP2000036104

    申请日：2000-02-15

    Applicant: MOTOROLA INC

    Inventor： POPPERT DANIEL CHARLES

    IPC: G10L15/02

    Abstract: PROBLEM TO BE SOLVED: To provide a feature vector extracting method used for a speech recognition system. SOLUTION: This feature extracting method includes a proces where a representative coefficient from a digitizing signal is generated. A differential coefficient is generated from the representative coefficients (304, 305). The partial representative coefficient and differential coefficient are extracted for forming the feature vector (307). Memory capacity used for the system is reduced by the feature vector, and a processing speed is accelerated.

    Method of selectively assigning a penalty to a probability associated with a voice recognition system
    4.
    发明专利
    Method of selectively assigning a penalty to a probability associated with a voice recognition system 未知

    公开(公告)号：GB2347253B

    公开(公告)日：2001-03-07

    申请号：GB0003940

    申请日：2000-02-18

    Applicant: MOTOROLA INC

    Inventor： POPPERT DANIEL CHARLES

    IPC: G10L15/14

    Abstract: A voice recognition system (204, 206, 207, 208) assigns a penalty to a score in a voice recognition system. The system generates a lower threshold for the number of frames assigned to at least one state of at least one model and an upper threshold for the number of frames assigned to at least one state of at least one model. The system assigns an out of state transition penalty to an out of state transition score in an allocation assignment algorithm if the lower threshold has not been met. The out of state transition penalty is proportional to the number of frames that the dwell time is below the lower threshold. A self loop penalty is applied to a self loop score if the upper threshold number of frames assigned to a state has been exceeded. The out of state transition penalty is proportional to the number of frames that the dwell time is above the upper threshold.

    Method for Channel Scanning
    5.
    发明专利
    Method for Channel Scanning 未知

    公开(公告)号：CA2163937A1

    公开(公告)日：1996-06-15

    申请号：CA2163937

    申请日：1995-11-28

    Applicant: MOTOROLA INC

    Inventor： SCHELLINGER MICHAEL JOHN ,  POPPERT DANIEL CHARLES

    IPC: H04B7/26 ,  H04B17/00 ,  H04Q7/36 ,  H04Q7/38

    Abstract: A unique method improves channel scanning by reducing the time required to scan channels. The method reduces the number of channels which must be scanned using information previously obtained in a scan. If two sets of channels which overlap are scanned in two separate scanning steps (522), information obtained in a first channel scan can be used, thereby eliminating the time required for a second channel scan. For example, if the dedicated control channels and the paging channels of a cellular telephone system overlap, the channels which overlap need not be scanned during the channel scan of the paging channels. The method further reduces scanning time by ending the channel scanning of a particular channel if an undesired channel type is detected (808).

    Method of extracting features in a voice recognition system
    7.
    发明专利
    Method of extracting features in a voice recognition system 未知

    公开(公告)号：GB2347775A

    公开(公告)日：2000-09-13

    申请号：GB0003949

    申请日：2000-02-18

    Applicant: MOTOROLA INC

    Inventor： POPPERT DANIEL CHARLES

    IPC: G10L15/02

    Abstract: A method of extracting features for a voice recognition system includes generating representative coefficients from a digitized signal (304). Differential coefficients are generated from the representative coefficients (305). A portion of the representative coefficients and the differential coefficients are extracted to construct a feature vector (307).

Patent Agency Ranking