Batoto sings
#21
Posted 08 February 2012 - 04:39 AM
#22
Posted 08 February 2012 - 05:21 AM
Unless there are any objections, I think we should proceed with Renai Circulation. I have an off vocal version of it (It's .flac if anyone cares/understands), so all we need are the voice samples. Audacity should easily be enough to do the job if there are no instruments to add.
Also, some advice to help make your voice sample better (Forgive me if this is obvious/mistaken, but saying it and having it be useless is better than not saying it when you should have): Sing along with the on vocal (it will improve your timing, which is absolutely crucial for a chorus), but only record your voice (it should do that anyways). I recommend Audacity to anyone who doesn't know any better program (It's pretty easy to just record your voice, you don't have to worry about anything else). And last and quite important, leave a couple seconds of silent space in your recording at the beginning or end (This will be useful for noise reduction).
EDIT: If you're okay with me doing the editing, you can send me samples at [email protected].
Edited by UnmeiNoShugosha, 08 February 2012 - 05:34 AM.
#23
Posted 08 February 2012 - 05:39 AM
Edited by UnmeiNoShugosha, 08 February 2012 - 05:51 AM.
#24
Posted 08 February 2012 - 11:57 AM
I'm tired as hell atm. I'm busy all the time. For that, I apologize if this post seems... strange compared to previous posts.
I'm going to sing. My range is shit, let's just leave it at that. Also, I'd be willing to bet that most of the people here don't know what their range is. I suggest we split voices into groups that harmonize well together. From there, we can then segment the song into sections and have each group sing a different part. Perhaps even solos for some. At one point, we might want to have all the groups sing together or something as well. I know I'm talking vague, but that general idea usually works well.
Seeing as we don't have a vocal coach for people, I suggest you just try your best. Take multiple recordings if you have to. Sometimes, you can sing small sections of a song separately, then splice it together to make it sound as if it was done all in one go.
That means, if you have multiple recordings under the same conditions, you might be able to splice the best parts of them together. I say this because some of the lyrics are a little tricky for English speakers. You'll probably mess up a lot and have to re-record.
.flac? I've never been a big fan of .flac, but whatever works.Unless there are any objections, I think we should proceed with Renai Circulation. I have an off vocal version of it (It's .flac if anyone cares/understands), so all we need are the voice samples. Audacity should easily be enough to do the job if there are no instruments to add.
UnmeiNoShugosha, prepare for tons of crappy singers (at least a lot of us claim to be) sending you samples. I hope you know how to enhance audio to make these vocals sound better.
Oh, yeah. For convenience, here's the written lyrics in romaji. Here's a video to sing along to that provides the lyrics. Practice, practice, practice.
Edited by Thunder Wolf, 08 February 2012 - 11:58 AM.
#25
Posted 08 February 2012 - 01:26 PM
i'll send my voice sample later after work. i have low range voice which would make me a bass i guess.
last questions : how many female:male ratio we have so far? not wanting to sound rude but ren'ai circulation with mostly male singing seems really wrong. we should have at least 3 or 4 girl. should we keep the limit on the members to 7 or 8 max? that way Shugo-san won't be having too much problem overlapping sounds.
#26
Posted 08 February 2012 - 01:36 PM
Seeing as this is the first song, I'll just be happy if we get anything out. The hardest part of any endeavor is starting it. Once we've got made something, more people will come.Wolf, when you're not so tired anymore why not update the front page. make a list on those taking part in the project so it'll be easier to keep track. we should also ask for another confirmation from members who have already said they want in.
i'll send my voice sample later after work. i have low range voice which would make me a bass i guess.
last questions : how many female:male ratio we have so far? not wanting to sound rude but ren'ai circulation with mostly male singing seems really wrong. we should have at least 3 or 4 girl. should we keep the limit on the members to 7 or 8 max? that way Shugo-san won't be having too much problem overlapping sounds.
Regarding the first post, I'll update it now. Regarding Renai Circulation, we have 2 dude and a girl singing it: katzen, Xinhope and I. I'll also update more important info for quick finding.
#27
Posted 08 February 2012 - 07:59 PM
.flac? I've never been a big fan of .flac, but whatever works.Unless there are any objections, I think we should proceed with Renai Circulation. I have an off vocal version of it (It's .flac if anyone cares/understands), so all we need are the voice samples. Audacity should easily be enough to do the job if there are no instruments to add.
UnmeiNoShugosha, prepare for tons of crappy singers (at least a lot of us claim to be) sending you samples. I hope you know how to enhance audio to make these vocals sound better.
The final product will be mp3, but .flac is a great starting point, since it will give the highest end-quality. I'm treating this as a project as "for fun," so I'm not worried about voice quality. As for enhancement, I'm pretty sure that other than noise reduction, the rest would require professional talent. I'll research to see if there's anything else I can do to improve the singing quality.
i'll send my voice sample later after work. i have low range voice which would make me a bass i guess.
This is a good example of what I meant by range. We have to know who is low or high, to mix them well. I'll know when I hear them, but this will help us plan in the forums.
Edited by UnmeiNoShugosha, 08 February 2012 - 08:00 PM.
#28
Posted 09 February 2012 - 12:45 AM
Also, I'd suggest adjusting the bass a little. That often makes things louder. I know loud doesn't equal good, but it can help in certain cases. Sometimes, Compressor + Normalize also make stuff sound better. Here's a joke example of how to do it. That link is not only informative, but funny as well.
- UnmeiNoShugosha likes this
#29
Posted 09 February 2012 - 01:04 AM
I'ma check out my exact range within a day or two. I don't remember exactly what it is, so I'll have to test it out.
Also, I'd suggest adjusting the bass a little. That often makes things louder. I know loud doesn't equal good, but it can help in certain cases. Sometimes, Compressor + Normalize also make stuff sound better. Here's a joke example of how to do it. That link is not only informative, but funny as well.
hahal XD The video cracked me up.
I do a pretty crappy female vocal but I'm willing to do it if there is lacking of female participants for Renai.
Thanks Shugo-san for helping out =)
#30
Posted 09 February 2012 - 01:29 AM
On a side note, a male Renai circulation isn't that bad. I've heard some really good male versions of it. In fact, I know 3 examples of that.
You might be thinking, "but he has a high range. Lower vocals can't do it". To which I reply with a video of Norio Wakamoto singing Renai Circulation.
http://www.youtube.com/watch?v=j7pyxpJFj18
That's right, it's really him.
For more convenience, here's a few more covers of Renai Circulation you might like.
Shizuo from Durarara!!
Yagami Light: Death Note remix
and Finally, The Pokedex Pidgey remix.
Those are just favorites of mine. If you look around, you can find a Rei Ayanami, Pikachu, Broly (BDZ), Seto Kaiba, Haruhi, Liechtenstein (Hetalia), Shin-chan and Gen (Gintama) version. There's probably even more.
Edited by Thunder Wolf, 09 February 2012 - 01:30 AM.
#31
Posted 09 February 2012 - 05:47 AM
I'ma check out my exact range within a day or two. I don't remember exactly what it is, so I'll have to test it out.
Also, I'd suggest adjusting the bass a little. That often makes things louder. I know loud doesn't equal good, but it can help in certain cases. Sometimes, Compressor + Normalize also make stuff sound better. Here's a joke example of how to do it. That link is not only informative, but funny as well.
If I do that, everyone will sound like Morgan Freeman!?!
EDIT: LIES!! I did it to my own voice and it made me sound like George Takei! Jk, but I tried out the features, and I have concluded that Compressor + Normalize will make everyone's voice's nice and even with each other, while making them clear as well, so I will use those and noise removal for certain. I might fool with the bass boost experimentally, as it does seem to improve the voice (probably only on males, though). None of the other features will be of much use. Thanks for the info, Wolf!
I also realized that if you act the role, then lower your voice and increase the bass (-4 half-steps, +10 Bass-boost for me) you can give yourself a sexy man voice... It was hilarious how well it worked for me.
Also, does anyone find it weird that I like listening to the off vocal version of this song because of how it brings out the tuba and bell parts?
Edited by UnmeiNoShugosha, 09 February 2012 - 06:29 AM.
#32
Posted 09 February 2012 - 08:00 AM
For reference: (s) means soloist. (g) means a group of 2, maybe 3 singers, if we get more people. (e) means everybody. These indicate that both the singer changes and how many there are. They don't refer to anyone in particular.
(#), however, refers to a specific soloist.
I like this setup because it mostly follows the balance of voices in the original, which I think works well. Also because the "rap" sections would probably be really hard for people to sing with enough synchronization to make piecing it together possible (rap is the most dependent singing on rhythm, and I know what happens when people playing the same rhythm are slightly off, as a drumline veteran. Bass line - shoes in a dryer, Snares and tenors - popcorn). Though it is fully possible.
Any questions/suggestions/improvements/completely different ideas?
Edited by UnmeiNoShugosha, 09 February 2012 - 08:03 AM.
#33
Posted 09 February 2012 - 02:09 PM
Changing pitch can also help a bit. For instance, my normal speaking voice is made slightly better by changing it using these settings: From C up to C#/Db
You can also toy around with tempo and speed. I suggest you don't change the speed that much, as it can make the voice too deep, or too high. Just use it as a makeshift bass/treble fix if you want.
After speeding it up or slowing it down, to re-sync it, you'd use tempo. Change Tempo basically makes the audio recording go fast or slower, but without altering how the voice sounds.
Amplify SOMETIMES makes things sound nicer. But you also have to watch for things like clipping. It can also sometimes make things too loud, so use with discretion.
Tremolo effect is good for making bad tremolo sound good. I just tested it out on my own voice.
I'll post a really crappy edit I did to make me sound like a vocaloid. I just said the word, "blah", then edited from there. It would have been better, but I'm tired and busy as usual. It's pretty bad.
Here's my recording
Here's what it sounds like after edit
Sorry for it not being a stream. I wasn't aware that .flac was hated by 4shared.
Tremolo -> Compressor -> Normalize -> Equalization (15dB to -9dB which didn't do much) -> Change Speed (faster) -> Phaser (I forget my exact settings) -> Echo (0.05) -> Tremolo -> Compressor -> Normalize -> Noise removal -> Change tempo (down to make my note longer)
On a side note, I think that arrange is good, but perhaps it can be improved.
Perhaps certain parts can be changed to have one person singing the first line, then having the next person in the group joining in? I'll use Group 1 = "1". So S1 = Solo from a person in Group 1.
(S1)Demo sonnanja dame
(S1)Mou sonnanja hora
(G1)Kokoro wa shinka suru yo motto motto
(S2)Sou sonnanja yada
(S2)Nee sonnanja mada
(G2)Watashi no koto mitetene zutto zutto
To clarify, person A from group 1 sings, then person B from group 1 sings, then they sing together. Does that sound good to you?
P.S. you can link youtube videos easier by hyperlinking. If you wanted to link to 2 minutes 4 seconds, you'd add &t=2m4s to the url.
Example
#34
Posted 09 February 2012 - 04:05 PM
anyway, before starting to talk about which person sings which part shouldn't we check each other voices first? everyone should do the full song once first and send it to shugo-san (and to all members maybe?) after that we can judge who can sing some parts and do some adjustments.
#35
Posted 09 February 2012 - 04:15 PM
яαιηвσω яσ¢кѕ ❤ヾ(◍’౪`◍)ノ゙|
#36
Posted 09 February 2012 - 04:35 PM
As for right now, I'm going to get someone to help me out with finding out how I should sing this. Glad I have musicians in my family. XD
#37
Posted 09 February 2012 - 06:29 PM
For females, you might want to adjust the treble. That makes their voice higher. But remember not to over do it.
Changing pitch can also help a bit. For instance, my normal speaking voice is made slightly better by changing it using these settings: From C up to C#/Db
You can also toy around with tempo and speed. I suggest you don't change the speed that much, as it can make the voice too deep, or too high. Just use it as a makeshift bass/treble fix if you want.
After speeding it up or slowing it down, to re-sync it, you'd use tempo. Change Tempo basically makes the audio recording go fast or slower, but without altering how the voice sounds.
Amplify SOMETIMES makes things sound nicer. But you also have to watch for things like clipping. It can also sometimes make things too loud, so use with discretion.
Tremolo effect is good for making bad tremolo sound good. I just tested it out on my own voice.
I'll post a really crappy edit I did to make me sound like a vocaloid. I just said the word, "blah", then edited from there. It would have been better, but I'm tired and busy as usual. It's pretty bad.
Here's my recording
Here's what it sounds like after edit
Sorry for it not being a stream. I wasn't aware that .flac was hated by 4shared.
Tremolo -> Compressor -> Normalize -> Equalization (15dB to -9dB which didn't do much) -> Change Speed (faster) -> Phaser (I forget my exact settings) -> Echo (0.05) -> Tremolo -> Compressor -> Normalize -> Noise removal -> Change tempo (down to make my note longer)
On a side note, I think that arrange is good, but perhaps it can be improved.
Perhaps certain parts can be changed to have one person singing the first line, then having the next person in the group joining in? I'll use Group 1 = "1". So S1 = Solo from a person in Group 1.
(S1)Demo sonnanja dame
(S1)Mou sonnanja hora
(G1)Kokoro wa shinka suru yo motto motto
(S2)Sou sonnanja yada
(S2)Nee sonnanja mada
(G2)Watashi no koto mitetene zutto zutto
To clarify, person A from group 1 sings, then person B from group 1 sings, then they sing together. Does that sound good to you?
P.S. you can link youtube videos easier by hyperlinking. If you wanted to link to 2 minutes 4 seconds, you'd add &t=2m4s to the url.
Example
I haven't noticed significant improvement with tremolo, but I'll see how it works on each voice and decide from there. I'll also manipulate treble as appropriate. Speed and pitch changes, however, are more likely to worsen singing than improve it. Amplify shouldn't be necessary.
I added your suggestions to the score(?). An updated version is below. Based on 4 vocalists. Working with the assumption that odds are female, evens are male.
Also, thanks for the tip on the link. I avoided linking the example because I didn't know how to set specific times.
EDIT: I will definitely be using tremolo.
Edited by UnmeiNoShugosha, 09 February 2012 - 07:15 PM.
#38
Posted 09 February 2012 - 06:52 PM
#39
Posted 09 February 2012 - 07:29 PM
That works pretty well. Thanks for taking the time to make that composition.
No problem, it was amazing fun. I've been practicing editing using my voice, and I have concluded that bass boost is definitely worthwhile for the guys, so I'll be using that, too. Dear god, I can't stop laughing after hearing my voice singing this.
#40
Posted 10 February 2012 - 04:25 AM
Thanks Shugo and Wolf for making the composition =)
Does anyone know if we sing the entire song or do we sing the assigned parts only?