Method for processing audio signals in a sub-band coding system

Claim 1

A method of processing audio signals in a sub-band coding system comprising the steps of: inputting non-linear audio signals; separating each non-linear audio signal into a plurality of frequency band signals; decoding each frequency band signal into a plurality of decoded linear signals; performing a predetermined computation on the decoded linear signals to generate computed linear signals; reproducing said non-linear audio signals by encoding and multiplexing the computed linear signals, said method characterized by including the steps of: selecting at least one decoded linear signal in which audio signal power is concentrated; detecting sound levels in the selected decoded linear signals; and applying the detected sound levels to the computed linear signals.

Claim 2

A method of processing audio signals in a sub-band coding system comprising the steps of: inputting non-linear audio signals; separating each non-linear audio signal into a plurality of frequency band signals; decoding each frequency band signal into a plurality of decoded linear signals; performing a predetermined computation on the decoded linear signals to generate computed linear signals; reproducing said non-linear audio signals by encoding and multiplexing the computed linear signals, said method characterized by including the steps of: selecting at least one decoded linear signal in which audio signal power is concentrated; detecting sound levels in the selected decoded linear signals; and applying the detected sound levels to the computed linear signals.

Claim 4

A method of processing audio signals in a sub-band coding system comprising the steps of: inputting non-linear audio signals; separating each non-linear audio signal into a plurality of frequency band signals; decoding each frequency band signal into a plurality of decoded linear signals; performing a predetermined computation on the decoded linear signals to generate computed linear signals; reproducing said non-linear audio signals by encoding and multiplexing the computed linear signals, said method characterized by including the steps of: selecting at least one decoded linear signal in which audio signal power is concentrated; determining a gain coefficient of the selected decoded linear signals; and controlling the level of the computed linear signals with the gain coefficient.

Claim 5

A method of processing audio signals in a sub-band coding system comprising the steps of: inputting non-linear audio signals; separating each non-linear audio signal into a plurality of frequency band signals; decoding each frequency band signal into a plurality of decoded linear signals; performing a predetermined computation on the decoded linear signals to generate computed linear signals; reproducing said non-linear audio signals by encoding and multiplexing the computed linear signals, said method characterized by including the steps of: selecting at least one decoded linear signal in which audio signal power is concentrated; determining a gain coefficient of the selected decoded linear signals; and controlling the level of the computed linear signals with the gain coefficient.

Claim 7

A method of processing audio signals in a sub-band coding system comprising the steps of: inputting non-linear audio signals; separating each non-linear audio signal into a plurality of frequency band signals; decoding each frequency band signal into a plurality of decoded linear signals; performing a predetermined computation on the decoded linear signals to generate computed linear signals; reproducing said non-linear audio signals by encoding and multiplexing the computed linear signals, said method characterized by including the steps of: selecting at least one decoded linear signal in which audio signal power is concentrated; detecting sound levels in the selected decoded linear signals and utilizing the sound levels to determine a gain coefficient; and controlling the level of the computed linear signals with the gain coefficient when the sound levels have been detected.

Claim 8

A method of processing audio signals in a sub-band coding system comprising the steps of: inputting non-linear audio signals; separating each non-linear audio signal into a plurality of frequency band signals; decoding each frequency band signal into a plurality of decoded linear signals; performing a predetermined computation on the decoded linear signals to generate computed linear signals; reproducing said non-linear audio signals by encoding and multiplexing the computed linear signals, said method characterized by including the steps of: selecting at least one decoded linear signal in which audio signal power is concentrated; detecting sound levels in the selected decoded linear signals and utilizing the sound levels to determine a gain coefficient; and controlling the level of the computed linear signals with the gain coefficient when the sound levels have been detected.

Claim 9

A method of processing audio signals in a sub-band coding system comprising the steps of: inputting non-linear audio signals; separating each non-linear audio signal into a plurality of frequency band signals; decoding each frequency band signal into a plurality of decoded linear signals; performing an N-1 addition process to the decoded linear signals to generate computed linear signals, the N-1 addition process for the purpose of holding an audio teleconference among at least two locations, where N is the number of conference participants; selecting at least one decoded linear signal in which audio signal power is concentrated; detecting sound levels in the selected decoded linear signals and utilizing the sound levels to determine a gain coefficient; controlling the level of the computed linear signals by applying the determined gain coefficient to the computed linear signals when the sound levels have been detected; and reproducing said non-linear audio signals by encoding and multiplexing the computed linear signals.

Claim 10

A method of processing audio signals in a sub-band coding system comprising the steps of: inputting non-linear audio signals; separating each non-linear audio signal into a plurality of frequency band signals; decoding each frequency band signal into a plurality of decoded linear signals; performing an N-1 addition process to the decoded linear signals to generate computed linear signals, the N-1 addition process for the purpose of holding an audio teleconference among at least two locations, where N is the number of conference participants; selecting at least one decoded linear signal in which audio signal power is concentrated; detecting sound levels in the selected decoded linear signals and utilizing the sound levels to determine a gain coefficient; controlling the level of the computed linear signals by applying the determined gain coefficient to the computed linear signals when the sound levels have been detected; and reproducing said non-linear audio signals by encoding and multiplexing the computed linear signals.

Claim 11

An apparatus for processing audio signals, comprising: means for receiving non-linear wide-band audio signals; means for separating each non-linear audio signal into a plurality of frequency band signals; means for decoding each frequency band signal into a plurality of decoded linear signals; means for performing a predetermined computation on the decoded linear signals to generate computed linear signals; means for reproducing the non-linear audio signals by encoding and multiplexing the computed linear signals, means for selecting at least one decoded linear signal in which audio signal power is concentrated; means for detecting sound levels in the selected decoded linear signals; and means for applying the detected sound levels to the computed linear signals.

Claim 12

An apparatus for processing audio signals, comprising: means for receiving non-linear wide-band audio signals; means for separating each non-linear audio signal into a plurality of frequency band signals; means for decoding each frequency band signal into a plurality of decoded linear signals; means for performing a predetermined computation on the decoded linear signals to generate computed linear signals; means for reproducing the non-linear audio signals by encoding and multiplexing the computed linear signals, means for selecting at least one decoded linear signal in which audio signal power is concentrated; means for detecting sound levels in the selected decoded linear signals; and means for applying the detected sound levels to the computed linear signals.

Claim 14

An apparatus for processing audio signals, comprising: means for receiving non-linear audio signals; means for separating each non-linear audio signal into a plurality of frequency band signals; means for decoding each frequency band signal into a plurality of decoded linear signals; means for performing a predetermined computation on the decoded linear signals to generate computed linear signals; means for reproducing said non-linear audio signals by encoding and multiplexing the computed linear signals, means for selecting at least one decoded linear signals in which audio signal power is concentrated; means for detecting sound levels in the selected decoded linear signals and for generating a gain coefficient therefrom; and means for controlling the level of the computed linear signals with the gain coefficient when the sound levels have been detected.

Claim 15

An apparatus for processing audio signals, comprising: means for receiving non-linear audio signals; means for separating each non-linear audio signal into a plurality of frequency band signals; means for decoding each frequency band signal into a plurality of decoded linear signals; means for performing a predetermined computation on the decoded linear signals to generate computed linear signals; means for reproducing said non-linear audio signals by encoding and multiplexing the computed linear signals, means for selecting at least one decoded linear signals in which audio signal power is concentrated; means for detecting sound levels in the selected decoded linear signals and for generating a gain coefficient therefrom; and means for controlling the level of the computed linear signals with the gain coefficient when the sound levels have been detected.