公开(公告)号：US3509281A

公开(公告)日：1970-04-28

申请号：US58291266

申请日：1966-09-29

Applicant: IBM

Inventor： KING JOHN H JR

IPC: G01R23/00 ,  G10L11/06 ,  G10L15/00 ,  G10L19/02 ,  G10L1/00

CPC classification number: G10L15/00 ,  G01R23/00 ,  G10L19/02 ,  G10L25/93

Abstract: 1,139,017. Speech recognition. INTERNATIONAL BUSINESS MACHINES CORP. 15 Sept., 1967 [29 Sept., 1966], No. 42164/67. Heading G4R. [Also in Division H4] In a speech analysis system, a signal representative of the harmonic energy of a speech signal is obtained from the sum of rectified signals each representing a respective harmonic content of the speech signal. Speech is split by filters at 3 having passbands of width 300 hz and centre frequencies as shown, the filter outputs being rectified at 4, passed through low-pass filters (15 hz) at 25 and summed at 29 to give the total spectral energy at 32. The outputs of filters 4 are also passed to 9 where they are summed, band-pass filtered (70 hz to 150 hz viz fundamental) at 13, fullwave rectified at 14 and low-pass filtered (15 hz) at 15, to give the harmonic energy at 16 (voiced sound). The energy signals at 32, 16 are compared at 17 to produce a voiced/unvoiced binary indication at 20d, the gains of the summing amplifiers 10, 30 being appropriately chosen for this purpose.

公开(公告)号：US3600516A

公开(公告)日：1971-08-17

申请号：US82941469

申请日：1969-06-02

Applicant: IBM

Inventor： KING JOHN H JR

IPC: G10L25/90 ,  G10L25/93 ,  G10L1/04

CPC classification number: G10L25/93

Abstract: Voicing detection and pitch extraction from speech sounds are achieved by means of an embodiment including a plurality of bandpass filters each having sufficient passing bandwidth to pass at least two harmonics of the fundamental voice frequency, whereby each provides a signal for all voiced sounds in the form of modulated waves, the envelopes of which having a periodicity equal to the voice fundamental. This periodicity is further enhanced by means of a hard limiter. A frequency discriminator whose input is provided by the band-pass filtered output of the limiter provides a voltage waveform whose special energy distribution is utilized for discrimination between voiced and unvoiced sounds.

公开(公告)号：US3446950A

公开(公告)日：1969-05-27

申请号：US3446950D

申请日：1963-12-31

Applicant: IBM

Inventor： KING JOHN H JR ,  TUNIS CYRIL J ,  MARCUS MITCHELL P

IPC: G06K9/66 ,  G06G7/28 ,  G06F7/02 ,  G11B13/04

CPC classification number: G06K9/66

公开(公告)号：US3600056A

公开(公告)日：1971-08-17

申请号：US3600056D

申请日：1969-11-24

Applicant: IBM

Inventor： KING JOHN H JR

IPC: G03H1/00 ,  G03H1/20 ,  G03H1/26 ,  G02B27/00

CPC classification number: G03H1/00 ,  G03H1/20 ,  G03H1/26

Abstract: High fidelity holograms can be replicated in quantity from a master hologram constructed by side band techniques in which spatial filters are employed in both reference and object beams. A collimating lens provides a plane wave reference beam which interferes with a convergent object beam produced by a focusing lens at an image plane occupied by the master hologram, the object function being located either before or after the focusing lens. In the replication process, the master hologram containing a plurality of discrete holograms and the hologram plate on which replicas of the master are to be recorded are placed one on either side of the focusing lens and separated from the latter by equal distances. A reference beam is directed through the master, which yields a diffracted beam constituting the master image that passes through the focusing lens and interferes with a component reference beam to form interference patterns, characterized as different from those on the master holoarray, on the emulsion side of the plate, which when developed forms the replicated hologram.

公开(公告)号：US3499989A

公开(公告)日：1970-03-10

申请号：US3499989D

申请日：1967-09-14

Applicant: IBM

Inventor： COTTERMAN HOWARD W ,  KING JOHN H JR

IPC: G10L19/02 ,  G10L1/00

CPC classification number: G10L19/02

Abstract: 1,245,414. Speech analysis. INTERNATIONAL BUSINESS MACHINES CORP. 10 Sept., 1968 [14 Sept., 1967], No. 42905/68. Heading G4R. Sound analysing apparatus comprises a plurality of filter systems, each of which has an output of amplitude and phase corresponding to a respective frequency band of an input sound signal, a threshold system for each filter system for determining which filter system output is above a predetermined threshold, and control means for varying this threshold independently of the instantaneous value of an input sound signal. A speech waveform at 1 is split into frequency bands by phase filters 2a-n, each of which produces an output the phase of which depends on the difference between the frequency of the input and the centre frequency of the filter, the outputs being limited at 4a-n and compared in quadrant detectors 6a-n with the original speech waveform limited at 8. Each quadrant detector produces a 2-state output, the upper state occurring when both inputs exceed a threshold. The quadrant detector outputs, after low-pass filtering at 8a-n for averaging, set respective threshold latches 14a-n if they, added to a ramp voltage from ramp generator 12, exceed a threshold. Each threshold latch 14a-n which is set, sets a corresponding formant latch 19a-n, via logic circuitry 16a-n, 17a-n, provided the two adjacent threshold latches (if any) are not set. The number of formant latches set is counted by a counter 22. A predetermined count resets the ramp. A sample clock 10 produces timing signals, the start of each of which permits the ramp to start, and the end of each of which resets the ramp if it has not already been reset by the counter 22. The counter 22 and threshold latches 14a-n are reset at the end of each timing signal.