According to Table 6 , it can be seen that for the Whisper-based ATC speech recognition model, the recognition accuracy of the speech recognition model shows a decreasing trend with the deepening of the degree of accent of the speech data, but the correlation is not large; for the PPASR-based ATC speech recognition model, the recognition accuracy of the speech recognition model shows a significant decreasing trend with the deepening of the degree of accent of the speech data, which is slightly larger than the correlation of the Whisper-based ATC speech recognition model.
← all excerpts
Assessment and analysis of accents in air traffic control speech: a fusion of deep learning and information theory.
1
—
—