It used to be that everyone listened to music in their homes using hi-fi stereo systems with huge amps and speakers. But today, almost all music is listened to using ear-bud speakers.

Stereo sound through headphones often doesn't sound the way it should, since the recording is based on two sources, not on two destinations.

Given the digital source and the processing power available, it shouldn't be that difficult to convert a stereo source to close to what its binaural recording would have been.

Do mobile phones etc. that people use for music have a binaural setting, or even have that as the default mode? (And if not, why not?)


Some answers are saying that it simply isn't possible. I don't know why they say that.

Stereo recordings are made by using two or more microphones and mixing the result into two channels. When played through two speakers, someone listening at the sweet-spot in front of those speakers should hear a close approximation to what the original performance was.

Binaural recordings are made using a special dual microphone that records what each ear of a human listener would hear. When played through headphones, the listener will hear what they would have heard at the live performance.

It's possible to play a stereo recording through two speakers, position a binaural microphone at the sweet-spot, and produce a binaural recording of what someone listening to the stereo sound would hear. — Binaural re-recording - Wikipedia

Today's hardware and software is fast and powerful enough that it should be able to simulate that re-recording process.

Obviously it won't be as good as an original binaural recording, but an app should be able to convert the standard stereo recordings everyone listens to into a much more realistic binaural sound when using earbuds.

I'm asking why this isn't already done. Simply stating that it isn't possible is not a good answer. Perhaps it isn't possible, but if so, I'd like to know why.

  • 1
    You cannot generate binaural data from stereo. It's simply not possible. I'll let someone smarter than me fill in the answer as to exactly why, but expect to see references to Haas or the precedence effect – Tetsujin Sep 22 '19 at 16:14
  • "You cannot generate binaural data from stereo. It's simply not possible." But you can. One crude way would be to play the source through stereo speakers and record the result with a binaural microphone. That same process could be digitally simulated in software. – Ray Butterworth Oct 23 '19 at 13:31
  • Binaural recordings don't need 'special microphones' at all. You can make binaural recordings with a pair of lavaliers if you like. You can also simulate binaural from stereo using delay lines. It's rough and ready, but it works. See Answer below. – Mark Oct 24 '19 at 10:21

No, they do not. It might be possible to cross-connect a short delay across the left and right channels in order to simulate two speakers in front of the head, but this is generally not done. The left channel - as supplied by the media file or stream is simply routed to the left ear-bud, and the right channel to the right ear-bud. Generally, no more complex than that.

It's possibly available with an app, but generally the standard functionality would leave this sort of thing alone as it is the cleanest way to treat the signal.

If BL=Binaural Left and BR=Binaural Right...

  • BL=(Stereo Left) + (Delayed/filtered Stereo Right)
  • BR=(Stereo Right) + (Delayed/filtered Stereo Left)

The delay must be consistent and non-zero. Tune to taste. Slight low-pass filter across some of the higher frequencies as well.

The result will be a forward-perceived binaural stereo sound-stage and will only work on headphones. It's rough and ready, and not perfect, but it proves a concept.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.