Compression in audio: lowering sampling rate, lowering bit depth, reducing number of channels

Before everything starts, there is a question for everyone: Why do you need to compress a file? You are confused and decide to ignore this question for the time being and download something. After opening "Net Depression Cloud", you find a few songs you like in the daily recommendations. You are so excited that you feel you have met true love, so you immediately click to download and want to take these songs for yourself. In the download options, you find that there are several options: highest sound quality, very high sound quality, and normal sound quality. You think you must take care of your ears when listening to songs, so you choose the highest sound quality. As a result, "Net Depression Cloud" asks you to buy a membership, so you are angry... and choose the very high sound quality.

Recalling the question just now, you suddenly wonder if ultra-high quality sound is the highest quality sound with some compression? Because the file size of ultra-high quality sound is smaller than that of the highest quality sound. The answer is of course yes. Although the sound quality of ultra-high quality sound is not as nice as that of the highest quality sound, smaller files take up less disk space and are more convenient for data transmission, so compression technology is very important for audio. Of course, compression technology can also tempt people to spend money...

Now let's take a look at the compression method in audio. First, let's look at the formula of file size:

File size = duration * sampling rate * bit depth * number of channels

Since the duration of an audio cannot be changed, we can only start by reducing the sampling rate/bit depth/number of channels.

Let's first look at the sampling rate. The higher the sampling rate, the better the quality of an audio clip. Commonly used sampling rates are shown in the figure:

You can see that the lowest sampling rate is 11025Hz, which is used for voice and AM radio. You can see that the sampling rate of FM radio is twice that of AM, which is why we usually choose FM xxx instead of AM xxx when listening to the radio in our private cars, because FM has better sound quality. If you are interested, you can listen to the two types of radio next time you are in the car and compare them.

Another way to compress audio is to reduce the bit depth. Common bit depths are 8-bit and 16-bit. Compressing a 16-bit audio file of about 10MB to 8-bit can reduce the file size by about 5MB. Generally speaking, the bit depth of ordinary speech (with no high requirements for sound quality) is 8-bit, while for music with higher requirements for sound quality, the bit depth is usually 16-bit. After all, no one wants to listen to a song full of noise.

So what is the number of channels? Generally speaking, stereophonic audio is a method of sound reproduction, usually with at least two audio channels, which can create a feeling of "I hear the sound coming from two directions!", making the audio three-dimensional and closer to natural audio. For a stereo with two audio channels, the size of the audio file will be halved by reducing one channel. However, this operation will damage the sound, so it is only suitable for short sound effects or voices, and is not suitable for use in music.

In addition, audio files are not suitable for lossless compression, because consecutive audio samples with the same value are not common. Common audio compression methods are lossy compression, such as mp3. But mp3 provides a good compression rate while maintaining the quality of the audio.