Is it possible to intercept all audio data sent to a speaker and write it to a file? I am trying to extract the audio content of a video but I do not have a microphone.
With either a file or a streaming address, you can use ffmpeg to extract the audio.
I've used a Matroska (MKV) file because it can handle pretty much any audio codec - the only major exception I know of is Opus, and that's unlikely to be part of a video file at the moment. If you know which audio codec is in use, you can use a different format. If you just want to extract it as an MP3, you can re-encode:
Now, to answer your actual question, again with ffmpeg. First, use the following command to list your dshow devices:
...the last few lines of which should be something like this:
In my case, I used (on Windows 8)
...which captured all the sound from my laptop's audio output. The name of the relevant device will probably be different on your computer. To stop recording, press