Reprinted from: https://blog.csdn.net/Marenow/article/details/85253283
Take notes, write down some of their own summary of the basics of audio.
Sampling rate
outside sounds are analog signals, A / D converted into a digital signal in a digital device represented by 0, after storage down. Digital signals are discrete, the sampling rate is the number of a second sample, the higher the sampling rate, the more real to restore sound. Since the human auditory range of 20Hz ~ 20kHz, according to Shannon's sampling theorem (also known as the Nyquist sampling theorem), in theory, the sampling rate is greater than 40kHz audio format can be called lossless format. But obtained at 40kHz sampling rate speech has no detail at all, all frequencies are only a sampling of a peak trough. The sampling frequency is now generally professional equipment is 44.1kHz. 44.1kHz sampling rate is the lowest in the professional audio, also called "CD-quality audio" (22.05kHz broadcast-quality audio sampling rate). There are more refined 96kHz, 192kHz, and so, of course, to hear the details of these higher sampling rate depends on the ear and the equipment.
Bit Depth
To restore the sound as accurately as possible, only the high sampling rate is not enough. Describes a sample point, and the horizontal axis (time) represents the sampling rate, and the vertical axis (amplitude) on behalf of the bit depth. 16bit represented by 16 bits (2 bytes) indicates the level of the sampling points (popular point, proportional to the size and volume) can be achieved when the encoding degree of accuracy, i.e. the longitudinal axis is divided into 16 parts description size level, and the accuracy of such difference -3.1415926dB of -3dB. Similarly there are 20bit and 24bit. 16bit is considered to be inside the field of professional audio bit depth minimum standards, and sample rate of 44.1kHz as common as the standard professional audio and consumer products. Bit depth also directly related to the magnitude of the signal to noise ratio, a direct impact on the overall dynamic range of the signal being recorded.
Rate
lossless uncompressed format (e.g., .wav), rate = sample rate x number of channels bit depth x. In lossy compression (such as .mp3) rate will not equal this formula, because the original information has been destroyed. It describes the audio bit rate of a second amount of information, and thus the total size of the sound file rate x is the total length. Rate also called bit rate in units of bit rate (bps, bit per second). Usually when the songs of 128kbps, 320kbps bit rate are, which is the highest bit rate of 320kbps mp3 format. However, 44.1kHz and the sampling rate, 16bit bit depth compared wav files (calculate two channel code rate is 44.1x16x2 = 1411.2kbps), far. After compression rate will be changed. Lossless compression rate has nothing to do with sound quality, lossy compression bit rates and sound quality a positive correlation.
Lossless compression
Lossless compression refers to the compression between lossless formats (conversion), whether compressed (converted) into what format, the sound quality is the same, and the same can be reduced to the original file. Usually said means are lossless lossless compression, lossless rate no argument. For a variety of formats are compressed corresponding to an algorithm (or coding), playback time requires a decoder for decoding, and different decoders may affect the integrity of the file unzipped. Common lossless formats:
WAV: a sound file format of Microsoft, is closest to the real sound uncompressed format (followed by midi), to support multi-rate multi-quantization precision. All formats are essentially lossless compression wav, will be back when playing wav.
flac: Free Lossless Audio Coded, is an international format, features high compression ratio, the encoding algorithm is quite mature, when the flac file damage still play properly. In addition, the format is a lossless format was first extensive hardware support.
ape: Use Monkey's Audio software for CD ripping and convert file format, but the advantage is not prominent, decoding slower.
wma-lossless: Microsoft also produced, is characterized by high compression ratio, but did not become mainstream.
aiff: Apple produced, is Apple Apple above standard audio format.
DSD: Sony Dafa, is not very understanding, appreciation is not to spicy kind of culture, but to say a simple punch, or the hub.
Lossy compression
Lossy compression means that audio information lost in the compression process took place, and the voice of the missing can not be represented by the sample rate and number of bits. However, after the file compression feature is very small, often used in media streaming. Common lossy formats:
MP3: simulation of the human auditory developed a complex algorithm, called "psycho-acoustic model." It is by extracting some of the audio band to achieve greater compression ratio, the lower rate, reducing the space occupied, but details such as the sound of the human voice emotions, the late reverberation, etc. have been deformed. Blind to hear it is difficult to quickly distinguish wav and mp3, needs the equipment. mp3 is currently the most popular audio compression format that can maximize the retention of sound before compression.
wma: Microsoft force, characterized by a low bit rate (e.g. 64kbps), wma smaller volume can be obtained under the same audio quality mp3. And ultra-low bit rate (e.g. 16kbps), wma much better sound quality than mp3.
aac: sound files stored on your computer Apple format.
ogg: completely free, open, and no patent restrictions, but popularity is poor.
If wrong, also hope you feel free to correct me!
----------------
Disclaimer: This article is the original article CSDN bloggers "Marenow", and follow CC 4.0 BY-SA copyright agreement, reproduced, please attach the original source link and this statement. .
Original link: https: //blog.csdn.net/Marenow/article/details/85253283