5 ms·

I thought this was going to be about adding spatial audio processing [1] to recorded music. One of the reasons recorded music is uncanny is that the act of list

by sxp 5y ago

I thought this was going to be about adding spatial audio processing [1] to recorded music. One of the reasons recorded music is uncanny is that the act of listening to something with stereo channels piped directly into your ears is vastly different than listening to the same music in a room as it is affected by the room's dynamics. There is a lot of work being done in VR research labs to simulate realistic audio by using the correct transfer function when playing the audio track. If done properly, listening to prerecorded audio of a concert would be indistinguishable from a live concert. But doing it correctly requires dynamically adapting the audio stream for the user's head & ear shape and ear position with respect to the audio source. Synthetic spatial audio that passes a Turing test has been done in the lab but we probably won't see it in consumer devices for a few more years. VR headsets and Airpods are slowly getting there, but they don't simulate the user's ears. [1] https://smus.com/spatial-audio-web-vr/ https://smus.com/spatial-audio-web-vr/

slfnflctd 5y ago
> listening to prerecorded audio of a concert would be indistinguishable from a live concert It seems to me this would be really dependent on the type of music and the size of the venue. In a large percentage of the bigger shows I've been to, the sound was honestly pretty awful. That's not really what I was there for. I was there to see what the artist(s) could do unedited in real time, in person, and to be (somewhat) physically near them when it happened. It's a different skillset from what they do in recordings.
082349872349872 5y ago
At some point in the 80's or 90's I read a book by someone who was at AT&T and into simulating room dynamics. One thing I remember is that, according to him, a large amount of classical music sounds much better when the transfer function approximates a relatively tall, narrow room (like a cathedral) than when it approximates a relatively short, wide room (like a modern concert hall). (note that construction spans, lighting, and ventilation, all conspire to make historical rooms taller and narrower than our typical rooms)
- Slow_Hand 5y ago
  Well I don't know what the writer's exact reasoning is, but there are few factors for why larger rooms are favorable: 1) Room modes. Small rooms are far more destructive because they feature hot-spots where the sound waves (especially the low end) are either cancelled out or reinforced in a way that makes them too loud. Large ceilings and lots of the depth to the room give the low frequencies space to propagate completely and a more even frequency response that isn't lumpy or lacking when you stand in certain parts of the room. 2) Reverberation. Pretty obvious, but the average length of the echoes/reverberant tail increase as the room gets bigger. This is gives an ambient quality to the sound that is well suited to orchestral ensembles. You could probably argue that the history of orchestral composition grew in tandem with this quality and informed the choices that were made or play best with the qualities of a room large enough to house an orchestra and audience. That said, highly percussive music might be muddied or hard to articulate cleanly if it's fighting against a massive reverb tail in a cathedral.

by sxp 5y ago

slfnflctd 5y ago

> listening to prerecorded audio of a concert would be indistinguishable from a live concert It seems to me this would be really dependent on the type of music and the size of the venue. In a large percentage of the bigger shows I've been to, the sound was honestly pretty awful. That's not really what I was there for. I was there to see what the artist(s) could do unedited in real time, in person, and to be (somewhat) physically near them when it happened. It's a different skillset from what they do in recordings.

082349872349872 5y ago

At some point in the 80's or 90's I read a book by someone who was at AT&T and into simulating room dynamics. One thing I remember is that, according to him, a large amount of classical music sounds much better when the transfer function approximates a relatively tall, narrow room (like a cathedral) than when it approximates a relatively short, wide room (like a modern concert hall). (note that construction spans, lighting, and ventilation, all conspire to make historical rooms taller and narrower than our typical rooms)

Slow_Hand 5y ago
Well I don't know what the writer's exact reasoning is, but there are few factors for why larger rooms are favorable: 1) Room modes. Small rooms are far more destructive because they feature hot-spots where the sound waves (especially the low end) are either cancelled out or reinforced in a way that makes them too loud. Large ceilings and lots of the depth to the room give the low frequencies space to propagate completely and a more even frequency response that isn't lumpy or lacking when you stand in certain parts of the room. 2) Reverberation. Pretty obvious, but the average length of the echoes/reverberant tail increase as the room gets bigger. This is gives an ambient quality to the sound that is well suited to orchestral ensembles. You could probably argue that the history of orchestral composition grew in tandem with this quality and informed the choices that were made or play best with the qualities of a room large enough to house an orchestra and audience. That said, highly percussive music might be muddied or hard to articulate cleanly if it's fighting against a massive reverb tail in a cathedral.