Stoke Amateur Theatre Society 2023 Christmas Panto

by Charlotte Pollock

Documenting the sound editing process of a dragon roar...

Dragon EQ: Boosted around 100Hz-800Hz to bring out the main body (impact) of this particular sound. 

Then I triple-tracked the dragon, keeping one track centre, and panning the others left (L) and right (R) to create a larger stereo image (i.e. the perceived space of the sound) 

Music: Cut from 20Hz to around 70Hz as this is where the sub-bass lies, and the dragon roar is very bass-heavy, so to make the dragon stand out more over the music, I cut out these frequencies in the music to “leave space” for the roar. Also, I did a bit of a cut in the high frequencies around 15-20kHz to ensure no unwanted high-frequency noise occurs in the track. 

I used a sidechain compressor on the music to make the volume dip slightly when the roars happen so that the roar cuts through the music better, making it more impactful. A sidechain is when the dynamics (volume content) of a track is controlled by another track, e.g. the dragon track is connected (i.e. side-chained) to the compressor on the music track to control when the music track dips. [Compressor settings: Threshold -7dB, Ratio 2:1, Make Up Gain 0.5dB, Attack 20ms, Release 90ms approx.]

Threshold – the level at which the compressor will start to take effect. 

Ratio – how strong the compressor is (e.g. 2:1 means for every 2dB over the threshold, the compressor will drop the volume by 1dB).

Make Up Gain – used to “make up” for the overall volume that has been lost when compressing. 

Attack – the speed at which the compressor works. 

Release – the speed at which the compressor allows the track to go back to its original level once it stops compressing. 

Lastly, I added a stereo spreader to the dragon track just to give it a final treatment. A stereo spreader takes the different frequencies that make up a sound and spreads them around the stereo field to create more ‘space’, i.e. make it sound and feel larger/wider/deeper. The stereo field is the perceived space when listening to music, achieved by giving different recordings more volume or less volume in the left or right channels of speakers or headphones (panning does this too, but more precisely). 

Dialogue: This is just one example of the EQ on the dialogue tracks; every track requires slightly different EQ settings. This was the EQ on someone with a slightly more high-pitched voice, so I cut from around 20Hz-300Hz as these low frequencies aren’t really present in the voice (only ambient room noise would lie around this range), so this cleaned up the sound. I boosted around 4kHz to give the voice more presence. 

My advice for dialogue EQ would be to have a play around with boosting different frequency bands and listen to where the main voice content is, then give a little boost to enhance. Cut out unneeded frequencies, especially the very low-end (20Hz) and very high-end (20kHz) to clean up the sound and reduce any noise. 

Finally, I added small fades at the beginning and ends of each dialogue track to ensure no clicks or pops occurred. When cutting down audio recordings, it can result in a clicking or popping sound where the cut was made, so adding a little volume fade solves this quickly. 

Listen Here: