What bit depth should I use? What sample rate should I use? 192kHZ or 96kHZ?

Lots of people don’t know what’s the difference between bit depth and sample rate, they think the higher the better, but what are bit depth and sample rate really mean?

Bit Depth

First of all, we need to understand what is bit. Bit is the basic unit of information in the digital world. Bit depth is how many bits of information in each sample. Record with higher bit depth will make the waveform more accurate and have more dynamic range. A 16-bit recording has a dynamic range of 96db, while a 24-bit recording has a dynamic range of 144db.

           

Sample Rate

Sample rate is how many samples contained in an audio per second, so if the sample rate is 20kHZ, that’s mean for each second of audio, it contains 20k samples. If you sample a 20kHZ frequency using the sample rate of 20kHZ, you will only have one sample per completed cycle, that is, to capture both sides of the waveform, you will need to use the sample rate of 40kHZ. To get accurate frequency response, you will need the sample rate at least double the frequency you want to sample.

 .             

What Bit Depth and Sample Rate Should I Use

Bit depth x Sample rate = The number of bits per second. Just like the number of pixels in a picture, the higher the bits it contains, the higher the resolution, but also the higher the file size. The standard CD bit depth and sample rate is 16-bit, 44.1kHZ. The frequency range of human hearing is 20HZ to 20kHZ, so the sample rate of 44.1kHZ is enough. You don’t necessarily need to export your file higher than 16-bit, 44.1kHZ, but you will want to work with higher bit depth (24-bit is enough) because you can have more dynamic range, so how do I export higher bit depth to lower bit depth.

Dithering

When you need to export a file with higher bit depth to lower bit depth, you need to dither first, if you don’t do this, you will lose some information. Dithering is basically randomly adding noise to randomize quantization error. The noise we added is very hard to notice, but it can help us fix the quantization problem.