COMPRESSION OF THE SPEECH STREAM SPECTRUM BY RESAMPLING SOUND FILES
Rubrics: ARTICLES
Authors:
1.
Saint Petersburg State University of Aerospace Instrumentation
2.
Design and Construction Bureau «RIO»
3.
Emperor Alexander I St. Petersburg State Transport University
Type:
Article
Pages:
from 43
to 49
Status:
Published
Received:
11.01.2025
Accepted:
11.01.2025
Published:
11.01.2025
Subject area:
UDK УДК
Language:
Russian
Keywords:
sampling theorem, compression and decompression of the speech stream, sampling and resampling of the signal
Abstract (English):
In this paper, we consider the problem of the maximum possible compression of the digital speech stream for its subsequent transmission over a narrow-band communication channel while maintaining speech intelligibility after decompression processes are performed on the receiving side. In contrast to the known methods of sound compression using vocoder principles, it is proposed to use: compression of the speech spectrum by resampling; reduction of the dynamic range of the speech stream; transmission to the communication channel of not samples of the oversampled signal, but only significant real and imaginary components of it; packing 4-bit words with information about the real and imaginary components of the signal into a byte stream; Decompression of the received stream is performed in the reverse sequence.
In this paper, we consider the problem of the maximum possible compression of the digital speech stream for its subsequent transmission over a narrow-band communication channel while maintaining speech intelligibility after decompression processes are performed on the receiving side. In contrast to the known methods of sound compression using vocoder principles, it is proposed to use: compression of the speech spectrum by resampling; reduction of the dynamic range of the speech stream; transmission to the communication channel of not samples of the oversampled signal, but only significant real and imaginary components of it; packing 4-bit words with information about the real and imaginary components of the signal into a byte stream; Decompression of the received stream is performed in the reverse sequence.
Keywords:
sampling theorem, compression and decompression of the speech stream, sampling and resampling of the signal
sampling theorem, compression and decompression of the speech stream, sampling and resampling of the signal
1.
