The author faced challenges in matching the timbre and loudness of the 200 songs, which led to the creation of a program to band-split all the files and match their perceived loudness. The final mixing of songs was also done programmatically, using a program that takes start and end times of the regions to crossfade, and then adjusts the samplerate for the two tracks. The author used OpenAI's GPT4 to write a BASH script to perform the auto-EQ.
Key takeaways:
- The author has compiled a mix of 200 "futuristic" music recordings from the mid 1970's to the mid 1980's and posted it on Soundcloud.
- The compilation process took almost two years due to the technical challenge of matching the timbre and loudness of 200 songs.
- The author wrote a program to band-split all the files and match their perceived loudness, and used OpenAI's GPT4 to write a BASH script to perform the auto-EQ.
- The final mixing of songs was also done programmatically, using a program that takes start- and end- times of the regions to crossfade, and then ramps samplerate up or down for the two tracks.