Track for the final assignment of the Audio Signal Processing for Music Application course on coursera.com Description (excerpt from assignment): "the crowdsourced mutant electro tune" To make the track for this assignment, I used only natural/acoustic sounds from freesound.org and sms-tools transformations. no other samples or effects were used. This is a preview of the final mix of the track, which has a quite experimental, tongue-in-cheek industrial vibe A10 - track master For the rhythmic parts, I picked a double bass "col legno" sample: <a href="http://freesound.org/people/Carlos_Vaquero/sounds/153796/" rel="nofollow">http://freesound.org/people/Carlos_Vaquero/sounds/153796/</a> To make the task more challenging I decided to use different transformations to obtain all the percussive parts from this sample only. For the melodic/harmonic parts I instead decided to limit myself to voice samples. However, I adopted a different approach compared to the percussive sample. Instead of handpicking a single sample myself, I sent a query to freesound using the API and the python script employed in assignment A9. SD.downloadSoundsFreesound(queryText='voice',API_Key='', outputDir='melodies/',topNResults=20,duration=(0,13), tag = 'female , singing') This way I quickly downloaded 20 voice samples with 'female' and 'singing' tags. This is the result of the query: voice_SoundList I listened to all the sounds and discarded the heavily processed or synthesised ones and the choir samples, keeping only the samples with a single female voice. The samples I kept are the following: <a href="https://www.freesound.org/people/xserra/sounds/162043/" rel="nofollow">https://www.freesound.org/people/xserra/sounds/162043/</a> <a href="https://www.freesound.org/people/maurolupo/sounds/211623/" rel="nofollow">https://www.freesound.org/people/maurolupo/sounds/211623/</a> <a href="https://www.freesound.org/people/maurolupo/sounds/211628/" rel="nofollow">https://www.freesound.org/people/maurolupo/sounds/211628/</a> <a href="https://www.freesound.org/people/toiletrolltube/sounds/181317/" rel="nofollow">https://www.freesound.org/people/toiletrolltube/sounds/181317/</a> <a href="https://www.freesound.org/people/maurolupo/sounds/211961/" rel="nofollow">https://www.freesound.org/people/maurolupo/sounds/211961/</a> <a href="https://www.freesound.org/people/Shuinvy/sounds/257360/" rel="nofollow">https://www.freesound.org/people/Shuinvy/sounds/257360/</a> <a href="https://www.freesound.org/people/visual/sounds/320626/" rel="nofollow">https://www.freesound.org/people/visual/sounds/320626/</a> <a href="https://www.freesound.org/people/EdwardLin2015/sounds/327756/" rel="nofollow">https://www.freesound.org/people/EdwardLin2015/sounds/327756/</a> <a href="https://www.freesound.org/people/womb_affliction/sounds/330709/" rel="nofollow">https://www.freesound.org/people/womb_affliction/sounds/330709/</a> Then I started transforming and overlapping the sounds. All the percussive element where obtained from the same double bass bass samples which is both percussive ("col legno" articulation) and pitched (it's a G#), therefore is suitable for various transformations. I created a whole drum kit transforming the same sample in different ways beyond recognition. To obtain the transformed samples I used the HPS transformation. The voices were first used clean and then later in the piece transformed to obtain strange effects and sounds.