Data augmentation by frame insertion for speech data

A computer system for data augmentation for speech data, by executing program instructions, the computer system comprising:

• a memory storing the program instructions;
• processing circuitry in communications with the memory for executing the program instructions, wherein the processing circuitry is configured to:
• obtain original speech data including a sequence of feature frames, each having dynamic acoustic features that include delta features and delta-delta features;
• generate a partially prolonged copy of the original speech data by inserting one or more new frames into the sequence of feature frames, including calculation of updated values of the dynamic acoustic features for each neighboring frame around each inserted new frame, with the calculation for the delta-delta features being done for a group of neighboring frames wider than the delta features;
• and
• output the partially prolonged copy as augmented speech data for training an acoustic model.

A computer program product for data augmentation for speech data, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to perform a computer-implemented method comprising:

• obtaining original speech data including a sequence of feature frames, each having dynamic acoustic features that include delta features and delta-delta features;
• generating a partially prolonged copy of the original speech data by inserting one or more new frames into the sequence of feature frames, including calculating updated values of the dynamic acoustic features for each neighboring frame around each inserted new frame, with the calculation for the delta-delta features being done for a group of neighboring frames wider than the delta features;
• and
• outputting the partially prolonged copy as augmented speech data for training an acoustic model.