posted 05 March 2012 18:01            

---

I am trying redo the effect of Starscream's voice here: http://www.youtube.com/watch?feature=player_detailpage&v=O6nt1JUuOOo#t=292s

(at 4:53)

I'm playing with noise+oscil vocoder but i can't get it quite right.

Anybody got another suggestion?
Thanks a lot.

IP: Logged

posted 05 March 2012 20:20            

---

starscreamsketch.kym

It seems like it might be a mix of vocoded plus a formant shifted voice? Something like the attached?

[This message has been edited by SSC (edited 05 March 2012).]

IP: Logged

posted 06 March 2012 09:58            

---

That sounds Great!
Thanks, i really appreciate it.

IP: Logged

posted 06 March 2012 10:43            

---

One small issue.
I need to batch process many files.
But when i add the DiskRecorder-SetDuration-Script at the end of the sound, i run out of static ram. even if i put oscillators to 1

Is there another way i could batch process, that would require less ram?

IP: Logged

posted 06 March 2012 11:51            

---

Wonder if the Script is overlapping the copies of the Sounds? Could you please attach the Script Sound you're using? Thanks!

IP: Logged

posted 06 March 2012 12:49            

---

starscreamsketch.kym

Here it is.
By the way. Would it be possible to have a fader for the noisiness, like in the noise+oscil vocoder?

IP: Logged

posted 06 March 2012 16:55            

---

starscreamsketch2.kym

Hmmm... I tried it here and did not run into the SRAM limit. Are you using a Capybara or Pacarana? How many files (approximately) are you processing?

I switched off the AutoLooping on the GenericSource and added some time to the end of each recording (since there is a Delay in the patch). And also added some time between recordings to allow for pre-fetching from disk.

To get the Gain control back, I changed the VCS to the top layout (using the control over on the right), so that all controls were visible again. Gain is controlling the noisy bits.

IP: Logged

posted 06 March 2012 17:39            

---

Allright, the sram issue is gone :P

Only one issue, it seems it take like a second after the sound has started for the vocoder to appear in the sound.
(alpha_male starts then vocoder).
And all processed files are like that.
Maybe the prefetching is not long enough?

IP: Logged

posted 06 March 2012 18:28            

---

You can line them up by adjusting !Delay and then saving it as part of the preset.

IP: Logged

posted 06 March 2012 21:17            

---

The two (vocoder+alphaMale) are already in-sync in the preset i use.
But the sound of the vocoder only appears a second after the sound has been started.
It appears abruptly. This is not just the batch sound, but sound alone also.
(i have a capy with 4 cards btw)

IP: Logged

posted 07 March 2012 09:37            

---

alien_andro.wav

Here is an audio example of what i mean

IP: Logged

posted 07 March 2012 15:51            

---

Working on this, will have an answer soon.

IP: Logged

posted 08 March 2012 10:39            

---

starscreamsketchRev3.kym

The variable delay module, when set for pre-zeroing mutes its output for the duration listed in the Delay field of that module. This is what causes that part of the effect to start later than the other part.

We changed the signal flow so that the GenericSource module is routed through a 1 second fixed delay before proceeding to the rest of the patch. This makes it so that the variable delay is no longer being muted when the audio to be processed begins.

IP: Logged

posted 08 March 2012 16:21            

---

Perfect!
Thank You for your help!

IP: Logged

All times are CT (US)