recently, I was messing around in the control panel to try to check my speakers. I managed to find the audio sample rate, which shows options from 44100 hz to 192000hz. there were also audio bit depth, but I don’t know either of those. what is audio sample rate? and maybe also what is audio bit depth?
In: 2
So, an analog sound wave is [continuous, like this](https://media.monolithicpower.com/wysiwyg/1_33.png). Analog waves, including sound waves, are converted to digital in two steps:
1) “sample” the wave every X milliseconds or microseconds. Let’s say X=1ms, it means: take the value of the analog wave at t=0, t=1 ms, t=2ms, t=3ms and so on. This is the concept of sampling frequency, that is measured in Hz. 44100 Hz it means you take 44100 values per second, i.e. every 22-23 microseconds.
2) represent this value (which could be something like 0.3423483234892) in a digital way, using N bits -> this number of bits is the audio bit depth.
The sampling frequency relates to which frequencies are faithfully represented by the digital version. The higher the sampling frequency, the wider is the range of sound frequencies that are correctly translated into digital.
The bit depth relates to how accurate the amplitude (“volume”, more or less) is.
Audio is analog. If a computer is involved, it has to be converted between analog and digital.
Imagine a audio curve going up and down.
The distance between the highest peak and lowest valley needs to be devided into distinct values.
The more values/bits are available, the closer it gets to the wave.
The sample rate is how to divide the wave horizontaly. 1hz would mean looking where the wave is every second. So the tone would only change ever, second. The more values/hz per second are used, the closer it gets to the wave.
Getting to the original analog wave would basically need infinitely many values in both axis.
You know how movies are 24 static images per second? Audio sample rate is the number of static sound fequencies per second.
Already some good answers on this, but it’s also important to understand the effects of different settings on these options.
Sampling rate is how often the audio signal is sampled. The rate you choose here will determine the highest frequency that your audio system can reproduce. Whatever rate you choose, your system will be limited in frequency to half that rate. So, if your sampling rate is 44.1k, the highest frequency it can reproduce is 22.05k. If sampling rate is 96k, then highest frequency is 48k, etc. Luckily, humans can’t hear over 20k, so most common sampling rate are ok. However, sometimes you’ll encounter weird sampling rates, like in digital telephone systems you might see 16k or 8k sampling rate. That’s generally ok because we’re used to telephone audio sounding pretty bad. Analog telephones don’t transmit much of anything above 3k.
Bit depth refers to how many “steps” you have to choose from between the lowest and highest voltage representing the audio wave. A lot of people think of this or describe it as the “resolution” of the sound, but in reality what this setting determines is how much noise your system will introduce into the sound. Technically it’s called quantization noise. If you constrained a sound to 4 bits, you’d hear tons of crazy noise. At 16 bit, noise is generally inaudible unless you really are pumping up the sound a lot (like playing it into a basketball arena at full volume, but even then, the crowd noise would drown out any tiny amounts of noise). At 24+ bits, noise is extremely low.
A sampling rate of 44.1k or 48k and a bit depth of 16 bits is generally sufficient for the vast majority of listening environments. Higher settings are often used in pre- and post-production environments, recording studios, and professional broadcast or live sound applications, for good reason. While audio is being processed over and over again, it’s better to use higher settings. But for a typical listening environment, lower settings are totally fine.
Latest Answers