All Day’s Child - first time out using a vocoder

edited February 2022 in Creations

Inspired by the classic nursery rhyme, Monday’s Child. A wee bit out there at times.

Produced entirely within NS2.
Synths used: just Obsidian and Slate with one track of Pure Synth. Spoken voices courtesy of Siri.
Vocoder is DerVoco. Not a lot of it though. I fed it Obsidian synth tracks as carriers and obsidian sampler tracks as modulators. External effects: bark filter, shimmer verb, black hole, squashit and FAC transient

Hope you enjoy.


  • Cool track! Nice range of musicality and some creative vocals make this a unique track. Nicely done. Did you bring the vocal audio clips into Slate?

  • Thanks @SlapHappy ! The text reading yes. The vocal lines that went into vocoder I brought into obsidian sampler. That allows for mono legato. I.e. the sample does not retrigger on new notes if you overlap them. You can have the carrier in either legato or poly. If the carrier is legato you can glide between notes without interrupting the phrase. If the carrier is in poly you can play chords and still not interrupt the modulator phrase as long as the modulator is legato. Learned a lot doing this. Would still be easier if I could sing :/

  • That sounds like it was quite the learning experience. Your explanation makes sense. I’m assuming the notes played on Obsidian (sample OSC) were recorded and could be adjusted in the Piano Roll Editor if there was need. This could be fun to play around with at some point.
    Singing… yeah…. Agreed.

  • Your assumption is correct. There is a single midi track (the “none” track discussed in the other thread) that hold the notes for both carrier and modulator. Midi track sends to carrier and modulator tracks. Don’t output the audio of these to main. The audio from those is sent to another “none” track with vocoder as audio effect. You need to prerecord the modulator audio and play the carrier, via the midi track, along with the prerecorded modulator. You can fine tune the tempo of modulator playback by transposing it in obsidian. I learned that recording the modulator works best if you “sing” the words at a single pitch (doesn’t matter which)- or as close as you can get to that when you can’t sing. I found that spoken word pitch inflections don’t produce ideal results in the vocoder. A deesser helps too.

  • Interesting! I would have thought singing as close to the desired pitch as possible would be best, but I guess the consistency of a single pitch would be easier to process afterward. Just curiois about your workflow, did you record audio directly into NS2 and de-ess using one of the iOS apps you listed or did you record audio externally and de-ess (in Audacity or something) prior to bringing the audio into NS2? I see there are some de-ess apps available in the AppStore too.

  • I played around with putting it on just the modulator audio track vs the vocoder output. Not sure if one is better than the other, just different.

    I recorded directly into NS2 using the iPad microphone. Probably would sound better if I sang the correct pitches. But I wanted flexibility to use the same recorded sample with different notes and chords. Well that, and my singing voice these days is a weapon of mass destruction :)

  • @boomer , you always takes us on some exquisite audio journey👌 This time I definitely enjoyed that vocal cuts and how you used them;) it added kinda psychedelic vibe to it in contrast to those melodies. Nicely done👏👏👏

  • Thank you sir. As always your kind words are much appreciated, as is listening to your music.

Sign In or Register to comment.