Gain adjusting method and a gain adjusting device

• US 7,916,874 B2
• Filed: 06/07/2006
• Issued: 03/29/2011
• Est. Priority Date: 03/09/2006
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Tsuchinaga, Yoshiteru, Makiuchi, Takashi, Shirakawa, Miyuki, Suzuki, Masanao
• Primary Examiner(s): Kuntz; Curtis
• Assistant Examiner(s): PRITCHARD, JASMINE L
• Application Number:
  US11/447,994
• Publication Number:
  US 20070223716A1
• Time in Patent Office: 1,756 Days
• Field of Search: 381/312, 381/318, 381/321, 381/93, 381/94.1, 381/94.2, 381/94.3, 381/94.4, 381/94.8, 381/104, 381/107, 381/56, 381/58, 381/110, 704/200, 704203-208
• US Class Current: 381/56
• CPC Class Codes: G10L 19/26 Pre-filtering or post-filtering
• CPC Class Codes: G10L 21/0232 Processing in the frequency domain
• CPC Class Codes: G10L 21/0364 for improving intelligibility
• CPC Class Codes: H03G 5/005 of digital signals
• CPC Class Codes: H03G 5/165 Equalizers; Volume or gain control in limited frequency bands

Abstract

In a gain adjusting method and a gain adjusting device for adjusting gain of a processed voice signal that is obtained by signal processing of an input voice signal,

• • a masking power of the processed voice signal is computed, and
  • gain is adjusted for every frequency if the frequency is masked according to the masking power, such that a difference between the processed voice signal and the input voice signal where the frequency is not masked is canceled.

Claims

1. A gain adjusting method which is performed to adjust a gain of a processed voice signal by using a gain adjusting device, the gain adjusting method comprising:

carrying out, by a first orthogonal transformation unit of the gain adjusting device, orthogonal transformation of a processed voice signal, which processed voice signal is obtained by carrying out signal processing of an input voice signal in a time domain, into a first spectrum power in a frequency domain;

generating, by a first band dividing unit of the gain adjusting device, a first spectrum energy from the first spectrum power wherein the first spectrum power is divided into N bands, each band having a predetermined bandwidth, and the first spectrum energy is represented by a sum of the divided spectrum powers of the N bands;

carrying out, by a second orthogonal transformation unit of the gain adjusting device, orthogonal transformation of the input voice signal in the time domain into a second spectrum power in a frequency domain;

generating, by a second band dividing unit of the gain adjusting device, a second spectrum energy from the second spectrum power wherein the second spectrum power is divided into N bands, each band having the predetermined bandwidth, and the second spectrum energy is represented by a sum of the divided spectrum powers of the N bands;

computing, by an auditory perception analyzing unit of the gain adjusting device, a masking power of the processed voice signal for each of the N bands, based on the first spectrum energy;

selecting, by a spectrum selection unit of the gain adjusting device, from among the N bands, bands in which the first spectrum energy is below the corresponding masking power of the processed voice signal;

determining, by an adjustment rate determining unit of the gain adjusting device, an adjustment rate for each of the N bands, such that a difference between a gain-adjusted spectrum energy, obtained as a result of gain adjustment of the first spectrum energy for the selected bands, and the second spectrum energy is equal to zero;

adjusting, by a gain adjusting unit of the gain adjusting device, respective spectrum powers of the first spectrum energy for the selected bands in accordance with the adjustment rates determined for the selected bands to generate the gain-adjusted spectrum energy; and

carrying out, by a third orthogonal transformation unit of the gain adjusting device, inverse orthogonal transformation of the respective spectrum powers of the gain-adjusted spectrum energy in the frequency domain, into an output voice signal in a time domain, and outputting the output voice signal to an external device.

2. The gain adjusting method according to claim 1, wherein the input voice signal and the processed voice signal are in a frequency domain.

3. The gain adjusting method according to claim 2, wherein the adjusting adjusts the gain for each frequency band.

4. The gain adjusting method according to claim 2, wherein the adjusting adjusts the gain for each frequency spectrum.

5. A gain adjusting device, comprising:

a first orthogonal transformation unit to carry out orthogonal transformation of a processed voice signal, which processed voice signal is obtained by carrying out signal processing of an input voice signal in a time domain, into a first spectrum power in a frequency domain;

a first band dividing unit to generate a first spectrum energy from the first spectrum power wherein the first spectrum power is divided into N bands, each band having a predetermined bandwidth, and the first spectrum energy is represented by a sum of the divided spectrum powers of the N bands;

a second orthogonal transformation unit to carry out orthogonal transformation of the input voice signal in the time domain into a second spectrum power in a frequency domain;

a second band dividing unit to generate a second spectrum energy from the second spectrum power wherein the second spectrum power is divided into N bands, each band having the predetermined bandwidth, and the second spectrum energy is represented by a sum of the divided spectrum powers of the N bands;

an auditory perception analyzing unit to compute a masking power of the processed voice signal for each of the N bands, based on the first spectrum energy;

a spectrum selection unit to select, from among the N bands, bands in which the first spectrum energy is below the corresponding masking power of the processed voice signal;

an adjustment rate determining unit to determine an adjustment rate for each of the N bands, such that a difference between a gain-adjusted spectrum energy, obtained as a result of gain adjustment of the first spectrum energy for the selected bands, and the second spectrum energy is equal to zero;

a gain adjusting unit to adjust respective spectrum powers of the first spectrum energy for the selected bands in accordance with the adjustment rates determined for the selected bands to generate the gain-adjusted spectrum energy; and

a third orthogonal transformation unit to carry out inverse orthogonal transformation of the respective spectrum powers of the gain-adjusted spectrum energy in the frequency domain, into an output voice signal in a time domain, and output the output voice signal to an external device.

6. The gain adjusting device according to claim 5, wherein the input voice signal and the processed voice signal are in a frequency domain.

7. The gain adjusting device according to claim 6, wherein the gain adjusting unit is configured to adjust the gain for each frequency band.

8. The gain adjusting device according to claim 6, wherein the gain adjusting unit is configured to adjust the gain for each frequency spectrum.