Audio encoding compression rate
The compression ratio of audio encoding is determined by the bitrate at the time of encoding.
In the previous article (Sample Rate and Bitrate, Part 1 –Smile Engineering Blog), I mainly wrote about the original sound (PCM) bitrate, but this time I would like to write about the bitrate and the compression rate. coding…
Specifically, if you set a low bitrate, the compression ratio will be high, for example, if you save the file, its size will be small. As I wrote last time, the sound source bitrate (PCM) before compression is as follows.
Bit rate = sampling frequency (Hz) x number of quantization bits x number of channels
For example, a music CD is 44.1 kHz stereo and the bit rate is
Music CD bit rate: 44100 Hz x 16 bits x 2 channels (stereo) = 1411.2 kbps
If it is encoded with MP3, AAC , etc., for example 256 kbps, the compression rate (assuming the original sound is 100%) is about 18% and the file size is 1/5 or less .
Encode a music CD at 256kbps: 256kbps / 1411.2kbps = about 18%
If it is 4 minutes of music, the file size is as follows.
44.1 kHz sample rate recording 4 minute song file size
as the original sound 1411.2 kbps x 240 seconds = about 40.4 MB
Encode at 256 kbps 256 kbps x 240 seconds = about 7.3 MB (+ header)
These days, it’s not the time to get songs on CD, so it’s an old story…
If a song is 4 minutes long, you can save 16 songs on CD650MB with the original sound, but at 256 kbps as MP3 and AAC . If encoded, 89 songs can be recorded.
Recording with a sampling frequency of 44.1 kHz The number of songs that fit on a CD (650 MB) of a 4-minute song
The original sound (music CD) CD650MB / 40.4MB = about 16 songs
Encode at 256 kbps CD650MB / 7.3MB = about 89 songs
If you check the web, you will be able to hear and compare the sound quality due to the difference in bit rate. I think the condition is that everything is the same except the bitrate, but first of all there is a difference in sound quality depending on the original sound source’s sampling rate (PCM) and the number of quantization bits (the bit rate changes from the original sound ) . At the time of analog-to-digital conversion (ADC), the sound quality is determined by the conditions. No matter how high the bit rate is to encode a sound source in poor conditions, the sound quality will still be poor. Even with the same bitrate, the compression rate changes depending on the number of channels (stereo or monaural). Therefore, strictly speaking, the evaluation of sound quality cannot be judged by the difference in bitrate alone. For example, when 48 kHz and 44.1 kHz 16-bit PCM is encoded at 32 kbps to 320 kbps, the compression ratio is as follows.
16-bit PCM compression ratio (when the original sound is 100%)
48 kHz stereo encoded bit rate
(1536kbps) 48kHz monaural
(768kbps) 44.1kHz stereo
(1411.2kbps) 44.1kHz monaural
(705.6kbps)
320kbps 320 / 1536 = about 21% About 42% 320 / 1411.2 = about 23% About 45%
256kbps 256 / 1536 = about 17% About 33% 256 / 1411.2 = about 18% About 36%
192kbps 192 / 1536 = about 13% About 25% 192 / 1411.2 = about 14% About 27%
160kbps 160 / 1536 = about 10% About 21% 160 / 1411.2 = about 11% About 23%
128kbps 128 / 1536 = about 8% About 17% 128 / 1411.2 = about 9% About 18%
64kbps 64 / 1536 = about 4% About 8% 64 / 1411.2 = about 5% About 9%
32kbps 32 / 1536 = about 2% about 4% 32 / 1411.2 = about 2% about 5%
Comparison with the original sound
It’s a slightly twisted idea, but for example, which is closer to the original sound, stereo or monaural under the above conditions? Considering the compression rate, it’s the latter. Of course, stereo is superior to monaural in terms of expression, such as expressing depth of sound, so it makes sense to compare this to assess sound quality, but in encoding, compression is efficiently done using stereo. Since there are algorithms (M/S stereo and intensity stereo), the quality is not even half as monaural, and the compression is done efficiently by the amount of stereo.
