Adpcm In Speech Coding - MULTIMEDIA

ADPCM

ADPCM forms the heart of the ITU's speech compression standards G.721, G.723, G.726, and G.727. The differences among these standards involve the bitrate and some details of the algorithm. The default input is μ - law - coded PCM 16 - bit samples. Speech performance for ADPCM is such that the perceived quality of speech at 32 kbps is only slightly poorer than with the standard 64 kbps PCM transmission and is better than DPCM.

Wave form of the word "audio:" (a) speech sample, linear PCM at 8 kHz and 16 bits per sample; (b) speech sample, restored from G.721 - compressed audio at 4 bits per sample; (c): difference signal between (a) and (b)

Wave form of the word

Wave form of the word

Figure shows a 1 - second speech sample of a voice speaking the word "audio." In Figure, the audio signal is stored as linear PCM (as opposed to the default / i - law PCM) recorded at 8,000 samples per second, with 16 bits per sample. After compression

G.726 quantizer

G.726 quantizer

with ADPCM using standard G.721, the signal appears as in the above figure. Figure shows the difference between the actual and reconstructed, compressed signals. Although differences are apparent electronically between the two, the compressed and original signals arc perceptually very similar.


All rights reserved © 2018 Wisdom IT Services India Pvt. Ltd DMCA.com Protection Status

MULTIMEDIA Topics