# Using preprocessed stems

If you don't want to use machine learning and instead want to use stems you obtained manually, this is possible since Spacer 0.2. To do so, place your stems in a folder named stems and use the base_manual.sds Scene file with your stems folder as the input audio. For example:

python main.py /path/to/my/stems base_manual.sds

Your stems folder needs to contain the following files:

• bass.wav
• drums.wav
• guitar.wav
• other.wav
• piano.wav
• vocals.wav