Benjamin O'Brien

I posted this question on the FB group, but I'll try again here. I'd like to synthesis crowd noise. For example, a crowd watching something might first say "oh" and then "ah" (like a Wii-sports golf game). I created something with this in mind (posted here) but it is not particularly authentic. Any recommendations would be greatly appreciated.

ps. Please don't post regarding using samples / recordings. The goal is to only use synthesis techniques. Thanks!

online.zip

I posted this question on the FB group, but I'll try again here. I'd like to synthesis crowd noise. For example, a crowd watching something might first say "oh" and then "ah" (like a Wii-sports golf game). I created something with this in mind (posted here) but it is not particularly authentic. Any recommendations would be greatly appreciated.

Ben

ps. Please don't post regarding using samples / recordings. The goal is to only use synthesis techniques. Thanks!

Hi all:
I posted this question on the FB group, but I'll try again here. I'd like to synthesis crowd noise. For example, a crowd watching so

synthesizing-crowd-noise

- have some synthesizer for one voice (very vague, i know)

- create the "crowd" by layering 1000 of them using random settings for e.g. start time, overall lenght, pitch and gain (and later maybe stereo position, timbre, intonation course...)

i usually do it by rendering groups of 20 members and then later summing the samples. be assured it can be fun and enlightening with percussive sounds - but dont miss out trying a choral piece.

 - have some synthesizer for one voice (very vague, i know)

 - create the "crowd" by layering 1000 of them using random settings for e.g.  start time, overall lenght, pitch and gain (and later maybe stereo position, timbre, intonation course...)

Thanks for your response Roman. Did you look at the patch I posted? Here I have multiple voices where each has a different fundmental frequency / oscillator for the source signal that drives a reson~ object (4x), where each has a different cf / q specified by formant values. I am not sure this method would be much different then layering multiple voices. I have played with different enveloping within each voice, but it still doesn't sit right. Thanks for checking it out! -Ben

didtn really look at the patch. but for pseudo voices maybe pay some attention to the vibrato. this is one of the most important "features" what makes a synth choir sound human like. (and in addition here it helps to avoid flanging and phasing when crowding)

Synthesizing crowd noise