Coding device, coding method, decoding device, decoding method, and storage medium

A coding processing method executed by a coding device that codes sound data, the method comprising:

• a sampling residual data generating step of delaying sampling data inputted among PCM waveform of sound data, multiplying the delayed sampling data by a linear prediction coefficient, and generating sampling residual data by subtracting the multiplied data multiplied from the inputted sampling data;
• an effective bit length data generating step of calculating an effective bit length for the sampling residual data generated by the processing of the sampling residual data generating step, and generating effective bit length data indicating the effective bit length of the sampling residual data;
• a common effective bit length retrieving step of retrieving a maximum effective bit length as a common effective bit length, with each effective bit length of a predetermined number of sets of the sampling residual data as a processing target, for the effective bit length data generated by the processing of the effective bit length data generating step, and generating, as a predetermined number of common effective actual data, data in which a portion other than the common effective bit length is erased from each of the predetermined number of sets of the sampling residual data;
• and
• a configuration format arranging step of generating, as coded data, data in which the predetermined number of common effective actual data generated by the processing of the common effective bit length retrieving step and information indicating the common effective bit length are at least arranged in a predetermined configuration format.

A non-transitory computer readable storage medium having a program stored thereon that is executable to control a computer controlling a coding device that codes sound data to function as:

• a sampling residual data generator comprising a delay circuit for delaying sampling data inputted among PCM waveform of sound data, and a multiplier for multiplying the delayed sampling data by a linear prediction coefficient, the sampling residual data generator generating sampling residual data by subtracting the data multiplied by the multiplier from the inputted sampling data;
• an effective bit length data generator for calculating an effective bit length for the sampling residual data, and generating effective bit length data indicating the effective bit length of the sampling residual data;
• a common effective bit length retrieving unit for retrieving a maximum effective bit length as a common effective bit length, with each effective bit length of a predetermined number of sets of the sampling residual data as a processing target, for the effective bit length data generated by the effective bit length data generator, and generating, as a predetermined number of common effective actual data, data in which a portion other than the common effective bit length is erased from each of the predetermined number of sets of the sampling residual data;
• and
• a configuration format arranger for generating, as coded data, data in which the predetermined number of common effective actual data generated by the common effective bit length retrieving unit and information indicating the common effective bit length are at least arranged in a predetermined configuration format.

A decoding device, comprising:

• a memory that stores coded data which is data in which information of a plurality of common effective bits, information indicating a common effective bit length, and information indicating a prediction coefficient that constitute sampling residual data calculated by linear prediction processing using the prediction coefficient for each set of sampling data of waveform data for sound, are at least arranged in a predetermined configuration format;
• a configuration format analysis unit that analyzes the arrangement of the information of the plurality of common effective bits, the information indicating the common effective bit length, and the information indicating the prediction coefficient, from the configuration format of the coded data;
• a common effective bit extraction unit for obtaining information indicating the common effective bit length from an analysis result by the configuration format analysis unit, and extracting the information of the plurality of common effective bits, respectively based on the information indicating the common effective bit length;
• a residual waveform data decompression unit for decompressing residual waveform data by performing bit extension, which adds a portion other than the common effective bit length, to the information of the plurality of common effective bits extracted by the common effective bit extraction unit, respectively;
• and
• an inverse linear prediction processing unit for performing inverse linear prediction processing by the prediction coefficient specified from an analysis result by the configuration format analysis unit on the residual waveform data decompressed by the residual waveform data decompression unit, so as to decode waveform data for the sound.

A decoding method executed by a decoding device that decodes waveform data for sound from coded data which is data in which information of a plurality of common effective bits, information indicating a common effective bit length, and information indicating a prediction coefficient that constitute sampling residual data calculated by linear prediction processing using a prediction coefficient for each set of sampling data of waveform data for sound, are at least arranged in a predetermined configuration format, the method comprising:

• analyzing an arrangement of the information of the plurality of common effective bits, the information indicating the common effective bit length, and the information indicating the prediction coefficient, from the configuration format of the coded sound data;
• obtaining information indicating the common effective bit length from an analysis result from the step of analyzing, and extracting the information of the plurality of common effective bits respectively based on the information indicating the common effective bit length;
• decompressing residual waveform data by performing bit extension, which adds a portion other than the common effective bit length, to the information of the plurality of common effective bits, respectively, extracted in the step of extracting;
• and
• performing inverse linear prediction processing by the prediction coefficient specified from an analysis result from the step of analyzing on the residual waveform data decompressed by the step of decompressing, so as to decode waveform data for the sound.

A non-transitory computer readable storage medium having a program stored thereon for enabling a computer, which controls a decoding device that decodes waveform data for sound from coded data which is data in which information of a plurality of common effective bits, information indicating a common effective bit length, and information indicating a prediction coefficient that constitute sampling residual data calculated by linear prediction processing using the prediction coefficient for each set of sampling data of waveform data for sound, are at least arranged in a predetermined configuration format, to execute functions comprising:

• a configuration format analysis step of analyzing the arrangement of the information of the plurality of common effective bits, the information indicating the common effective bit length, and the information indicating the prediction coefficient from the configuration format of the coded sound data;
• a common effective bit extraction step of obtaining information indicating the common effective bit length from an analysis result from the configuration format analysis step, and extracting the information of the plurality of common effective bits, respectively, based on the information indicating the common effective bit length;
• a residual waveform data decompression step of decompressing residual waveform data by performing bit extension, which adds a portion other than the common effective bit length, to the information of the plurality of common effective bits, respectively, extracted in the common effective bit extraction step;
• and
• an inverse linear prediction processing step of performing inverse linear prediction processing by the prediction coefficient specified from an analysis result from the configuration format analysis unit, on the residual waveform data decompressed in the residual waveform data decompression step, so as to decode waveform data for the sound.