Maybe the article title seems a little unusual at first glance. After all, the recording is already going through the sound card. Just to be clear, we're talking about recording sound that's played through the sound card. For instance, how can we record sound from a web browser? If the app doesn't have recording built in, the only way to record is to capture the sound going to the speakers.
1 AUX Cable
Now that full-duplex sound cards are available, all users can record sound directly from their sound cards. At first, you had to connect the sound card output to the input with an auxiliary stereo cable with 3.5 mm jacks. This method worked with any sound card, but had two big issues.
Firstly, the sound quality wasn't great after double digital-to-analog conversion, and most sound cards use a cheap DAC with a low signal-to-noise ratio. On top of that, there's electrical interference on an unshielded cable, which means it doesn't meet modern requirements. It was pretty good when it first came out 20 years ago.
Secondly, you couldn't hear the sound while recording because the sound card's speaker output was taken up by the cable. Theoretically, you could fix this by installing a second sound card. Later on, this same solution was used, but at the software level.
2 Stereo Mix
As you know, sound recording programs don't work directly with the sound card. They use the capabilities provided by audio drivers. Sound card manufacturers added their own way to solve this problem to the drivers. It's called "Stereo Mix". All sounds passing through your sound card were mixed digitally and were available for recording. All this happened digitally, which guaranteed consistent sound quality.
Since each manufacturer had a different name for this feature, it was tricky to automatically choose the right settings. On top of that, the RIAA was quick to pressure manufacturers to remove the feature because there was a chance that copy protection could be circumvented.
Another drawback of Stereo Mix was that it could only be used with one sound device. If you use a USB microphone with a sound card, you can only record from one source.
3 Virtual Sound Card
Third-party software developers didn't just sit back and let it happen. They started creating their own original solutions. The main approach was to install a virtual sound card in the system that could capture sound. That is, they created what the official vendors didn't or weren't willing to put in their drivers.
The solution turned out to be very simple - the virtual device, instead of playing, simply wrote all audio streams to disk. However, the implementation was disappointing. Virtual sound devices did not implement all the required functions, but worked only with the proprietary application directly. This caused a lot of headaches for third-party software that thought it was dealing with a normal sound card.
4 Code Injection
Other developers used techniques similar to those used in computer viruses and began to intercept applications calls to audio drivers. To do this, the software module was loaded into the address space of the target application, like a web browser or media player. Does this ring a bell? Yes, this is precisely how spyware operates, stealing private data. As you can imagine, this approach raised some red flags with antiviruses, so it wasn't widely adopted. From what I recall, it was only effective on older versions of Skype, but it could also result in a blue screen of death.
5 Loopback Recording
Finally, Microsoft decided to rid us of this problem once and for all. The audio stream management technology in Windows Vista had a built-in feature that allowed you to redirect and intercept sound going to the sound card.
You can use the "Speakers" device to record anything played through it, no matter who made the sound card. This means it'll work just as well and keep up with any system updates. You can redirect audio streams to mix sound from different sources and control the volume for each one. This is what our sound card recorder uses to record everything you hear in all modern versions of Windows.
Are there any downsides to this solution? Of course, there are a few drawbacks. Microsoft is bound by the law, so recording audio that's protected from copying using DRM isn't an option. There are still a few technical limitations that prevent us from calling this solution ideal. But this is the best we have right now.