revnice1
Max Output Level: -69 dBFS
- Total Posts : 1082
- Joined: 2007/09/21 14:43:56
- Status: offline
Ideas for a 'robot voice'
I want to do spoken piece in a song that sounds like a test of the emergency broadcast system. Right now, I'm using a Telephone EQ and it sounds suitably stupid but some clarity is lost, it sounds very Lo Fi (and I'd rather have Hi Fi) and it tends to bring out my sibilance. Anyone got a better idea? Thanks - rev
|
The Maillard Reaction
Max Output Level: 0 dBFS
- Total Posts : 31918
- Joined: 2004/07/09 20:02:20
- Status: offline
🤖
post edited by Bash von Gitfiddle - 2018/10/17 00:28:32
|
stickman393
Max Output Level: -60 dBFS
- Total Posts : 1528
- Joined: 2003/11/07 18:35:26
- Status: offline
Re: Ideas for a 'robot voice'
2018/10/16 21:26:11
(permalink)
Hard normalization > Hard pitch correction > filter > distortion > flanging Adjust the last three to taste
|
revnice1
Max Output Level: -69 dBFS
- Total Posts : 1082
- Joined: 2007/09/21 14:43:56
- Status: offline
Re: Ideas for a 'robot voice'
2018/10/16 22:06:39
(permalink)
Stickman: Filter? What kind of filter?
|
Euthymia
Max Output Level: -86 dBFS
- Total Posts : 223
- Joined: 2016/10/18 05:19:00
- Status: offline
Re: Ideas for a 'robot voice'
2018/10/17 00:14:47
(permalink)
You should be able to nail "Emergency Broadcast" using the tools in ProChannel. Squash it using the track compressor and narrow the EQ to midrange honk. For "Robot Voice," I like TAL Vocoder.
-Erik ___________ 3.4 GHz i7-3770, 8G RAM, Win 7 64-bit2X PreSonus Firepods, Event 20/20bas, Alesis Monitor Ones, Alesis Point SevensCakewalk by BandLab, Mixcraft Pro Studio 8.5Warning: if you tell me my issue can be remedied by buying more RAM, an SSD, or a Waves plug-in, I will troll you pitiilessly
|
Kamikaze
Max Output Level: -45 dBFS
- Total Posts : 3013
- Joined: 2015/01/15 21:38:59
- Location: Da Nang, Vietnam
- Status: offline
Re: Ideas for a 'robot voice'
2018/10/17 00:19:32
(permalink)
☄ Helpfulby iRelevant 2018/10/17 16:10:43
I have Vocodex, between it and Melda's they are probably the most affordable and usable. There is also Tal Vocoder which is free https://tal-software.com/products/tal-vocoder I am also working on a robt voice or a kids song,but instead of me being the source I'm using a text to voice generator https://www.text2speech.org/Then running that through vocodex But you escription o what you re after sounds more like Stickmans, Telephone/PA voice, and I beeive he just means low pass filter to lower the sound quality a little.
|
stickman393
Max Output Level: -60 dBFS
- Total Posts : 1528
- Joined: 2003/11/07 18:35:26
- Status: offline
Re: Ideas for a 'robot voice'
2018/10/17 03:10:49
(permalink)
☄ Helpfulby Kamikaze 2018/10/17 04:19:36
|
BenMMusTech
Max Output Level: -49 dBFS
- Total Posts : 2606
- Joined: 2011/05/23 16:59:57
- Location: Warragul, Victoria-Australia
- Status: offline
Re: Ideas for a 'robot voice'
2018/10/17 03:59:31
(permalink)
Depending on what type of robot voice you're after - there are a number of ways to achieve said sound.
If you want a Dalek - sorry if you're not from Britian or Oz and don't know what a Dalek is...if so look it up, but a Dalek voice is created by using a ring modulator. Go to KVR - a website that has information about plugins. You should be able to get one for free. I use GR4.
If you want a more traditional type of robot voice then a vocoder would be the go - Wave's has one and you can pick it up for 30 bucks normally. I have this plug, and whilst it's not very intuitive to use...it is excellent still.
Then if you want something more esoteric, you could use a granular delay. Again, I use GR4 or Guitar Rig 4...the granular delay in the effects suite is excellent, but again tricky to use if you're just starting out. You can pick up GR5 - I think for about 90 bucks...but well worth it because it's the swiss army knife of effects.
Continuing down the granular track, you could also use a granular synth. I have Steinberg's Padshop...in reality Padshop is in fact a quantum representation of all time based effects - meaning in theory it can replicate every time based effect and more. Think delays and modulation. I like to think of granular synthesis as a bit like Willie Wonka's Glass Elevator. This is because a granular synth at its core is a sampler - but unlike a sampler, which can only go back and forwards - a granular synth can back, forward, left and right and any which way in between. Of course unless you're British you might not know Willie Wonka - his chocolate factory and of course the more obscure follow up book by Roald Dahl.
Finally, if you're reluctant in experimenting with above techniques...as someone has already said you could try a filter. An auto-wha maybe, fed into a distortion - maybe chuck on a leslie type speaker and some delay might do the trick. Again KVR will have a free leslie or invest in the 90 bucks for Guitar Rig 5 because it too has a leslie.
Hope that helps.
|
msmcleod
Max Output Level: -72 dBFS
- Total Posts : 920
- Joined: 2004/01/27 07:15:30
- Location: Scotland
- Status: offline
Re: Ideas for a 'robot voice'
2018/10/17 08:42:08
(permalink)
Mark McLeod Cakewalk by BL | ASUS P8B75-V, Intel I5 3570 16GB RAM Win 10 64 + Win 7 64/32 SSD HD's, Scarlett 18i20 / 6i6 | ASUS ROG GL552VW 16GB RAM Win 10 64 SSD HD's, Scarlett 2i2 | Behringer Truth B2030A / Edirol MA-5A | Mackie MCU + C4 + XT | 2 x BCF2000, Korg NanoKontrol Studio
|
Wayfarer
Max Output Level: -74 dBFS
- Total Posts : 831
- Joined: 2017/07/15 02:32:25
- Status: offline
Re: Ideas for a 'robot voice'
2018/10/20 23:18:45
(permalink)
If it doesn't have to be "on beat", then the easy way is simply to use a text to mp3 converter. Instant robot. There are several free ones. Just Google them. (Incidentally, if you have an old Kindle Keyboard [K3] unit, it has a built-in text to mp3 converter, so you could run a line out of the earphone jack and record it after loading a text file to the unit.) I have a book formatting service, and one thing I always encourage clients to do before submitting a book to me is to run the text through an mp3 converter and listen to the book being read back to them because certain grammar errors are easy to spot when the text is being read to you, whereas your eyes might gloss right over them otherwise. Bill
post edited by Wayfarer - 2018/10/21 04:11:09
|
mettelus
Max Output Level: -22 dBFS
- Total Posts : 5321
- Joined: 2005/08/05 03:19:25
- Location: Maryland, USA
- Status: offline
Re: Ideas for a 'robot voice'
2018/10/21 04:48:26
(permalink)
stickman393 Or bandpass actually
Telephones passed 300Hz-3.4KHz so bandpassing that is a good start. It should take care of most sibilance as well, but you can always adjust to taste. You may also want to run the same filter back-to-back, to make sure you drive the filtered frequencies down, depending on which EQ you use and how effective it is. I would get that range to your liking before passing it on into any additional processing.
ASUS ROG Maximus X Hero (Wi-Fi AC), i7-8700k, 16GB RAM, GTX-1070Ti, Win 10 Pro, Saffire PRO 24 DSP, A-300 PRO, plus numerous gadgets and gizmos that make or manipulate sound in some way.
|