A speech encoding method, apparatus and program wherein an input speech signal is divided into a plurality of frames each having a predetermined length, each of the frames is subdivided into a plurality of subframes, a predictive pitch period of a subframe in a to-be-encoded current frame is obtained by using pitch periods of at least two frames of the current frame and past and future frames with respect to the current frame; a pitch period of a subframe in the current frame is obtained by using the predictive pitch period, a relative pitch pattern codebook storing a plurality of relative pitch patterns representing fluctuations in pitch periods of a plurality of subframes is prepared, and a change in pitch period of plural subframes is expressed with one relative pitch pattern selected from the relative pitch pattern codebook.

Speech encoding method, apparatus and program

PROBLEM TO BE SOLVED: To detect a defect of a seal conveniently by picking up the image of the upper surface of a thermally fused outer cover by means of an infrared thermal image camera and making a decision whether a seal is defective or not based on the shape of the image and the analysis of data. SOLUTION: A synthetic resin molding member laminated on the lower surface is fused by heat generated an aluminum seal through high frequency oscillation to seal the doughnut type upper edge at the mouth of a container. The generated heat is transmitted the rear top surface of an outer cover to the outer upper surface. The heat distribution is picked up from above by means of an infrared thermal image camera and processed by an image inspection processor to determine the temperature distribution on the outer upper surface of the outer cover which is then displayed in color or subjected to histogram analysis and compared with that of a product sealed well. A decision is made that the seal is acceptable it they match each other otherwise a decision is made that the seal is rejectable and an abnormal value is detected. According to the method, 100% inspection of fusion seal can be made conveniently and easily and quality control can be performed quite efficiently and effectively.

Noncontact inspection method for frequency aluminum seal

The invention relates to a method for decoding digital audio data. The inventive method is used for detecting errors according to transmitted reference values, preferably scale factors. According to the inventive method, reference values of a frequency range are compared to preceding reference values of the same frequency range for producing a characteristic that is compared to a threshold value. Signals are displayed for indicating when the characteristic exceeds the predetermined threshold value. In another embodiment of the invention, a default value is entered in frequency ranges, wherein audio data is not transmitted, and as a result there is no characteristic produced for said frequency range.

Method for decoding digital audio data

PROBLEM TO BE SOLVED: To provide a method capable of simply and extremely easily performing total inspection on the quality of a fused seal when a container (bottle) made of a resin is filled with contents and an outer cap in which an aluminum seal (inner cap) is placed is mounted to the container and passed below a high-frequency sealer to heat the aluminum seal without contact and fuse it to the upper edge of the opening part of the container. SOLUTION: The upper surface of the cap is photographed by an infrared thermal image camera continuously after the high-frequency induction heating processing. On the basis of the shape of an obtained infrared image or a histogram obtained by processing a thermal image, the quality of the aluminum seal is determined.

High-frequency and non-contact method for inspecting aluminum seal

PROBLEM TO BE SOLVED: To provide a container for a spherical object capable of taking out therefrom a spherical content easily one by one.SOLUTION: There is provided a container 1 for a spherical object for storing a spherical content M. The container 1 for the spherical object has: a container body 2 including a storage space S for the content M, and a discharge port 4 having a larger diameter than the content M; and a film body 10 mounted on the discharge port 4, which includes an aperture 11 having a smaller aperture area than a sectional area at a largest-diameter part of the content M, and at least one slit 12 provided extendedly from the inner peripheral edge of the aperture 11 toward the radial direction outside, and capable of deforming the aperture 11 until the content M becomes passable therethrough.SELECTED DRAWING: Figure 1

Container for spherical object

PURPOSE:To improve the speech quality of a decoded speech by decoding information that is transmitted in a way that continuity is maintained between frames, while interpolating the information. CONSTITUTION:A 1st interpolator 10 is connected between a pitch information decoder 3 and a pulse train sound source 7 and a 2nd interpolator 11 is connected between a power information decoder 5 and a synthesizer 9. The 1st interpolator 10 continuously and smoothly vary intervals between pulses in a current frame outputted by the pitch information decoder 3 so that the continuity with the repetitive intervals of the pulse train in the current frame is maintained. The 2nd interpolator 11, on other hand, further divides the frame into small sections and varies the power in small-section units so that the continuity between the power in the current frame outputted by the power information decoder 5 and the power in the last frame is maintained.

Shigeru Hosoi

Papers

Speech decoding device