Method and apparatus for extracting feature representation, device, medium, and program product

A computer device, comprising a processor and a memory, the memory storing at least one program, the at least one program being loaded and executed by the processor to implement a method for extracting a feature representation from an audio signal, the method including:

• obtaining sample audio;
• performing feature extraction on the sample audio from a time domain dimension and a frequency domain dimension to obtain a sample time-frequency feature representation corresponding to the sample audio;
• performing frequency band segmentation on the sample time-frequency feature representation to obtain time-frequency sub-feature representations respectively corresponding to at least two frequency bands;
• and
• performing feature sequence relationship analysis on the time-frequency sub-feature representations respectively corresponding to the at least two frequency bands along the time domain dimension to obtain an analyzed feature representation;
• performing dimension transformation on the analyzed feature representation to obtain a first dimension-transformed feature representation, wherein the first dimension-transformed feature representation is a feature representation obtained by adjusting a direction of the time domain dimension in a time-frequency sub-feature representation in the first dimension-transformed feature representation;
• performing inter-frequency band relationship analysis on the time-frequency sub-feature representation in the first dimension-transformed feature representation along the frequency domain dimension to obtain an inter-frequency band relationship analysis result;
• and
• obtaining a target time-frequency feature representation based on the inter-frequency band relationship analysis result, wherein the target time-frequency feature representation is used for a downstream analysis processing task applied to the sample audio.

A non-transitory computer-readable storage medium, having at least one program stored therein, the at least one program being loaded and executed by a processor of a computer device to implement a method for extracting a feature representation from an audio signal, the method including:

• obtaining sample audio;
• performing feature extraction on the sample audio from a time domain dimension and a frequency domain dimension to obtain a sample time-frequency feature representation corresponding to the sample audio;
• performing frequency band segmentation on the sample time-frequency feature representation to obtain time-frequency sub-feature representations respectively corresponding to at least two frequency bands;
• and
• performing feature sequence relationship analysis on the time-frequency sub-feature representations respectively corresponding to the at least two frequency bands along the time domain dimension to obtain an analyzed feature representation;
• performing dimension transformation on the analyzed feature representation to obtain a first dimension-transformed feature representation, wherein the first dimension-transformed feature representation is a feature representation obtained by adjusting a direction of the time domain dimension in a time-frequency sub-feature representation in the first dimension-transformed feature representation;
• performing inter-frequency band relationship analysis on the time-frequency sub-feature representation in the first dimension-transformed feature representation along the frequency domain dimension to obtain an inter-frequency band relationship analysis result;
• and
• obtaining a target time-frequency feature representation based on the inter-frequency band relationship analysis result, wherein the target time-frequency feature representation is used for a downstream analysis processing task applied to the sample audio.