Noise reduction method and noise reduction apparatus

A noise reduction method, wherein the noise reduction method is applied to a keyboard of an integrated terminal device and comprises:

• obtaining a first voice source and a second voice source, wherein the first voice source is a fidelity voice source, the second voice source is an audio signal that comprises the first voice source and a noise signal, the noise signal comes from noise generated by vibration of the keyboard, and/or the noise signal comes from noise of an environment in which the integrated terminal device is located, the fidelity voice source is an unplayed audio signal output by a power amplifier of the integrated terminal device, and the noise signal comprises a resonance signal generated by the keyboard and a loudspeaker;
• determining the noise signal based on the first voice source and the second voice source;
• and
• performing noise reduction processing on the second voice source based on the noise signal, to obtain a voice source that undergoes the noise reduction processing;
• and
• the performing noise reduction processing on the second voice source based on the noise signal comprises:
• performing noise reduction processing on the second voice source through filtering based on the noise signal;
• and
• playing, by the loudspeaker of the integrated terminal device, the voice source that undergoes the noise reduction processing.

A noise reduction method, comprising:

• displaying a display interface of an integrated terminal device;
• receiving a first operation of a user on the display interface;
• and
• responding to the first operation, so that the integrated terminal device performs operations comprising:
• obtaining a first voice source and a second voice source, wherein the first voice source is a fidelity voice source, the second voice source is an audio signal that comprises the first voice source and a noise signal, the noise signal comes from noise generated by a keyboard of the integrated terminal device, and/or the noise signal comes from noise of an environment in which the integrated terminal device is located, the fidelity voice source is an unplayed audio signal output by a power amplifier of the integrated terminal device, and the noise signal comprises a resonance signal generated by the keyboard and a loudspeaker;
• obtaining the noise signal based on the first voice source and the second voice source;
• and
• performing noise reduction processing on the second voice source based on the noise signal, to determine a voice source that undergoes the noise reduction processing;
• and
• the performing noise reduction processing on the second voice source based on the noise signal comprises:
• performing noise reduction processing on the second voice source through filtering based on the noise signal;
• and
• playing, by the loudspeaker of the integrated terminal device, the voice source that undergoes the noise reduction processing.