中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

19 records (took 0.016 seconds)

    12.
    发明专利
    未知

    公开(公告)号：DE3883519D1

    公开(公告)日：1993-09-30

    申请号：DE3883519

    申请日：1988-03-08

    Applicant: IBM

    Inventor： BOTTAU FRANCOISE ,  GALAND CLAUDE ,  ROSSO MICHELE

    IPC: G10L19/06 ,  G10L19/12 ,  G10L9/14

    Abstract: The voice signal s(n) is filtered through a short-term predictive filter (13) tuned with PARCOR derived coefficients computed over a pre-emphasized s(n), said filter (13) providing a short-term residual r(n). Said r(n) signal is then processed through a first Code-Excited/Long-Term Predictive coder providing first couples of table address and gain data (k1, g1)'s. An error signal r min (n) is then derived by subtracting coded/decoded data from uncoded data. Then said error signal is processed through a second Code-Excited/ Long-Term Predictive coder providing second couples of data (k2, g2)'s. Full rate coding is achieved by multiplexing both couples (k1, g1)'s and (k2, g2)'s into a multi-rate frame; while switching to a lower rate is achieved through a mere delation of (g2, k2)'s from the full rate frame.

    13.
    发明专利
    未知

    公开(公告)号：DE69123579T2

    公开(公告)日：1997-06-12

    申请号：DE69123579

    申请日：1991-09-06

    Applicant: IBM

    Inventor： MENEZ JEAN ,  ROSSO MICHELE ,  SCOTTON PAOLO

    IPC: H03H15/00 ,  H03H17/00 ,  H03H21/00 ,  H04B3/23

    Abstract: The echo cancellation method includes starting with an arbitrary set of echo filter coefficients. The coefficients are first sequentially re-estimated using a fast converging method. These operations are repeated a predefined number of times, and then the coefficients estimation is switched to a low converging but more precise method. In addition, an adaptive threshold is being defined and readjusted and all filter coefficients lower than said threshold are being discarded. Finally, sampling time boundaries are adaptively defined based on non-discarded filter coefficients with reference to said threshold, and in such a way the final echo cancelation algorighm is adaptive to echo variations (backtracking algorithm).

    15.
    发明专利
    未知

    公开(公告)号：DE3783905D1

    公开(公告)日：1993-03-11

    申请号：DE3783905

    申请日：1987-03-05

    Applicant: IBM

    Inventor： GALAND CLAUDE ,  ROSSO MICHELE ,  LIETHOUDT THIERRY ,  ELIE PHILIPPE ,  LANCON EMMANUEL ,  CREPY HUBERT

    IPC: G01H3/04 ,  G10L19/08 ,  G10L9/14

    Abstract: The digital process includes setting signal dependent positive threshold (Th+) and negative threshold (Th-). Samples representative of the signal block are located and stored having magnitudes above and below the thresholds. Significant sign transitions are located within the stored samples. The number M' of stored samples between the consecutively located significant transitions is computed and a rough value M is computed as the mean value of M'. Autocorrelation zones are set about the multiples of M spaced peak samples and the considered block of signal samples is split into consecutive sub-blocks. A current sub-block of samples is autocorrelated with sub-blocks of samples the first of which is one sample of an autocorrelation zone.

    METHOD FOR DETECTING VOICE PRESENCE ON A COMMUNICATION LINE
    16.
    发明专利
    METHOD FOR DETECTING VOICE PRESENCE ON A COMMUNICATION LINE 未知

    公开(公告)号：CA2076606C

    公开(公告)日：1996-07-23

    申请号：CA2076606

    申请日：1992-08-21

    Applicant: IBM

    Inventor： ARNAUD CHARLES ,  ROSSO MICHELE

    IPC: H04M3/22 ,  G10L11/02 ,  H04J3/17 ,  H04M11/06 ,  G10L9/18

    Abstract: A method and apparatus allows to accurately detect voice presence on a digitized communication line even in the presence of tone signals. The method includes analyzing digitized communication input signal by current blocks of 20ms (step 20), to define the stationary or non stationary state of any current block (step 21). Then an analysis on the states of the M last 20ms blocks (step 22) allows to give the final decision on voice presence or not.

    18.
    发明专利
    未知

    公开(公告)号：DE3688980D1

    公开(公告)日：1993-10-07

    申请号：DE3688980

    申请日：1986-10-30

    Applicant: IBM

    Inventor： ROSSO MICHELE ,  GALAND CLAUDE

    IPC: G10L19/02 ,  G10L19/00 ,  H03M1/12 ,  H03M7/30 ,  H04B1/66 ,  H04B14/04

    Abstract: A process for multirate subband encoding voice signals. At least a portion of the original signal bandwidth is plit into p sub-bands the contents of which is to be coded using dynamic allocation of quantizing levels throughout the sub-bands. Let's assume the signal is to be coded for a set of coding rates R(1) Then the sub-band samples S(i,j) are quantized according to 1(i; R(K). And, finally the obtained quantized sub-band samples are embedded into a multirate frame.

Patent Agency Ranking