Abstract:
A sub-band adaptive differential pulse code modulation/encoding apparatus includes means (102, 103, 104, 105) having a predetermined asymmetric impulse response for receiving an audio signal and band-dividing the received audio signal into a predetermined number of sub-bands, so as to obtain a plurality of band-divided sub-band audio signal, a plurality of quantization means (110, 111, 112, 113) for quantizing the band-divided sub-band audio signals by a predetermined divider, and encoding means (115) for performing adaptive differential pulse code modulation/encoding of the quantized sub-band audio signals.
Abstract:
The spectrum of a PCM signal is separated into bands. A combination is designated out of the combinations of the reference band including the highest frequency and the other bands so that the designated combination has the highest correlation of spectrum distribution when one of the reference band and the other band is normalized. The spectrum of the same distribution as the spectrum distribution in the reference band of the designated combination is so scaled that the spectrum is along the function expressing the envelope, and the scaled spectrum is added to the higher-frequency side of the reference band to generate an output signal. The PCM signal is checked for a high-frequency part. Only if a high-frequency part is detected, a spectrum component is added and an output signal is generated. Thus a signal approximated to the original signal can be restored from one of a signal representing a signal having a suppressed spectrum component in part of the bands of the original signal and the signal representing the original signal that does not originally contain the spectrum component of the band and even from a signal containing both signals.
Abstract:
A method for transferring audio data and audio-related information includes: a generation step of generating second audio data from first audio data; a transmission step of transmitting the second audio data and audio-related information associated with the second audio data; and a reception step of receiving the second audio data and the audio-related information, wherein the audio-related information includes information representing a sampling frequency of the first audio data.
Abstract:
The spectrum of a PCM signal is separated into bands. A combination is designated out of the combinations of the reference band including the highest frequency and the other bands so that the designated combination has the highest correlation of spectrum distribution when one of the reference band and the other band is normalized. The spectrum of the same distribution as the spectrum distribution in the reference band of the designated combination is so scaled that the spectrum is along the function expressing the envelope, and the scaled spectrum is added to the higher-frequency side of the reference band to generate an output signal. The PCM signal is checked for a high-frequency part. Only if a high-frequency part is detected, a spectrum component is added and an output signal is generated. Thus a signal approximated to the original signal can be restored from one of a signal representing a signal having a suppressed spectrum component in part of the bands of the original signal and the signal representing the original signal that does not originally contain the spectrum component of the band and even from a signal containing both signals.
Abstract:
The spectrum of a PCM signal is divided into bands. Combinations of a reference band inclusive of a highest frequency band and another band, one of the reference band and other band being normalized, are checked to identify a combination having a highest spectrum distribution correlation. The spectrum having the same distribution as the spectrum distribution of the reference band contained in the identified combination is scaled along an envelope function and added to a higher frequency side than the reference band to generate an output signal. A presence/absence of high frequency components of a PCM signal is detected. Only if there are high frequency components, the spectrum components are added to generate an output signal. It is therefore possible to recover a signal approximate to the original signal from either an original signal with the spectrum components in some bands being suppressed or a signal representative of an original signal containing no spectrum components in the bands or from a signal combining these two signals.
Abstract:
A voice signal or a modem signal input into a transmitter is coded, and the coded signal is transmitted from the transmitter to a receiver through a digital transmission line. In this voice coding apparatus, when the input modem signal is a modem signal output from a high-speed modem, it is transmitted without low bit rate compression. By virtue of this constitution, the voice coding apparatus can realize high-speed modem transmission not involving a lowering in transmission speed.
Abstract:
A voice encoding/decoding apparatus has a data buffer holding voice data and encoded data, a decoding process unit which receives the voice data of one subframe from the data buffer to decodes the same and outputs obtained encoded data of one subframe to the data buffer, a decoding process unit which receives the encoded data of one subframe from the data buffer to decodes the same and outputs obtained voice data of one subframe to the data buffer, a frame synchronization detecting unit, a first clock generating unit providing timing of inputting/outputting the voice data to/from the data buffer, a second clock generating unit providing timing of inputting/outputting the encoded data to/from the data buffer, a word transfer information providing unit providing the number of words of the voice data and the number of words of the encoded data to be inputted/outputted to/from the data buffer, and information about positions of the voice data and the encoded data in the data buffer, and a control unit minimizing a time required from an inputting to an outputting of the data upon encoding and decoding the data of each subframe on the basis of the information provided by the word transfer information providing unit.
Abstract:
Method of calculating the transmission quality of a series of cascaded codecs. For each type of codec, a function is determined which is specific for the relationship between the Q-equivalent over a series of codecs of that same type and the number of codecs which said series comprises. In addition, the type sequence is determined of the series of codecs such as those cascaded in the transmission medium. The series is traversed in steps and, starting from the Q-equivalent for one codec of the type which occurs as the first in the series of codecs, the Q-equivalent is always calculated over the traversed part of the series by calculating, starting from the Q-equivalent calculated last, the next Q-equivalent under the control of the function which is specific for the type of the last codec in said part which has been traversed. The Q-equivalent finally calculated for the entire codec series is converted into an MOS value which is a measure of the subjective quality of the codec series.