PCM Audio Coding Part 3
Sampling frequency
The human frequency recognition range is 20HZ – 20000HZ. If 20,000 sound samples per second can be sampled, it can meet the needs of human ears during playback.
8000 Hz for telephone sampling.
A sample rate of 22050 is typically used.
44100 is already CD quality, and samples over 48000 are meaningless to the human ear
When decoding AAC (Advanced Audio Coding) audio with a sampling rate of 44.1 kHz, the decoding time of one frame should be controlled within 23.22 milliseconds. Usually it is a frame with 1024 sampling points.
Why do we have to talk about audio frames here?
The concept of audio frames is not as clear as that of video frames. Almost all video encoding formats can simply think of a frame as an encoded image. But the audio frame is related to the encoding format, which is implemented by each encoding standard. Because if it’s PCM (unencoded audio data), you don’t need the concept of frames at all, and it can be played according to the sample rate and sample precision. For example, for audio with a sample rate of 44.1 kHz and a sample precision of 16 bits, you can calculate that the bit rate is 44100 16 kbps and the audio data per second is fixed at 44100 16/8 bytes .
But we don’t want all the samples returned to us for processing, what we want is to return all the data sampled over a period of time. The audio box here is how much sample data is returned to us each time and in general 2048 sample data is returned.
So what is the size of 2048 sample data using 16-bit sampling bits for mono? 2048*16/8 = 4096 bytes.
Sampling bits
Each sampled data registers the amplitude, and the sampling precision depends on the size of the storage space (sampling bits):
1 byte (ie 8 bits) can only register 256 numbers, that is, only the amplitude can be divided into 256 levels
2 bytes (ie 16 bits) can be as small as 65536 numbers, which is already the CD standard;
4 bytes (i.e. 32 bits) can subdivide the amplitude into 4294967296 levels, which is really unnecessary
If it’s stereo, the samples are doubled and the file is almost doubled in size.
