Previous and Next posts:
Tuesday, April 1st, 2008


If you like karaoke singing and you already shared your passion with your on-line friends of a community, one fine day you would be interested in the idea to have fun singing a duet with one of your friends.

A duet is easy when you sing with your friends, sister or brother at home, in the same room, same PC… but what if you on-line friend is in Japan and you’re in California?

In this case you should be able to setup a…
Virtual Duet!


You and your friend should decide who does what, there are two Roles:

  • the Mixer
  • the Partner

The Mixer should be aware of all the tech-stuff, at least basic audio recording concepts and, in most cases, also know about Midi editing and software. It depends if you just like to sing with your friend or to make the final ‘product’ sound like a good semi-professional recording: of course the experience of the Mixer would make this difference. By the way the important thing is to have fun together.

The duet Partner should simply know how to record his/her vocal track and send it to the Mixer, that’s all.

General Procedure

Basically the procedure steps are:

  1. Mixer and Partner agree on the song to sing together so they find the background music (midi-file or other audio) and share this file, same file;
  2. a wise choice is to search for a Midi file, especially if the song is not so popular so that no audio karaoke (e.g. mp3/wav) is available; in this case they should also agree on the Key of this song (if the song is too high it is possible to lower the music of the desired amount of semitones). Also the Tempo should be changed but… too much stuff to mention;
  3. Mixer -once the choice is made- should start working on the background music with audio editing software, modifying accordingly the midifile (in case of mp3 or other audio file, not much to do), trying to make it sound good;
  4. Mixer sings his/her part and records the first vocal track (multitrack recording software is recommended);
  5. Mixer exports a composite of background music and vocals, let’s say in mp3 (stereo and at least 128 Kbps), from the audio Project containing all tracks. Not necessarily a final product even a ‘draft‘ file, the important is that the key and tempo of the music are the chosen ones they previously agreed upon;
  6. Mixer sends this draft audio file to the Partner (by email as attachment, instant messaging ’send file’, etc…);
  7. Partner starts recording his/her vocal track (the second one): also in this case it is good to use a multitrack software so Partner could put the background in track 1 and record a second track with his/her voice only;
  8. Partner exports his vocal track only as a good quality mp3 (stereo and 128 Kbps is ok), so: no background music, less background noise as possible, no heavy breath on the mic and no effects (reverb, echo, etc..) on the voice;
  9. Partner sends this audio file to the Mixer;
  10. Mixer now has all the tracks and should do his/her best to make the final exported audio file sound good, adding effects to the voices, manage the stereo position of each track, etc…;
  11. Share the final product, it depends on the purpose: generally the final step is simply to upload the final mp3 somewhere but in case the website, community or service requires you to record directly from the microphone you should let your sound output to act like a mic, so a simple playback of your final mp3 will simulate you singing on the mic.

15 Responses to “Virtual Duet Procedure”

  1. Howard Says:

    Hey, Heracleum.
    How are you doing? Nice pages! (I often come to your site to play a round of “SUDOKU”.)

    Thanks for the good advice here about picking the background music and the key, about the “partner” sending the “mixer” a clean voice-only file. Very good advice for us first-timers.

    I think “the devil is in the details” though, for some of us. For example, the part about letting the “sound output” act like a mic is easy with XP, but weird with Vista. And many people don’t seem to know about the “settings” panel in Midomi’s flash-based recorder, where “mic” or “sound output” must also be selected to match the PC’s settings. These are the kinds of questions I expect you will get from novices. Still, EXCELLENT “How-to” procedure. I learned a lot from your first 8 or so steps. Time will tell if more details are needed.

    May I ask you, what midi player do you use? I have nothing, so I can only play midis in the”native” key, although I know most people can raise or lower the pitch by the desired number of semi-tones. (My singing range is SO NARROW that this issue makes many midis useless for me.)

    Lu2007 and I are in the early stages of planning a duet. I’m practicing the words of a Brazilian song. The same key as the original artists will work in this case, so our half-baked plan is for me to send Lu my vocal and she will be the mixer. We may ask you for help if we are having troubles. (She is practicing the words to a Japanese song, too!)

  2. Heracleum Says:

    Hey Howard! Glad to hear you’re a Sudoku fan :)
    Ah yes, about details… I already promised to Dewi to add more in-depth explanation. Soon I’ll write separate posts and just add here links to them on the different steps, just to leave the basic procedure ’slim’ and quickly readable.
    Since I don’t know Vista, this is good to know, so let me check if I got it right: while in XP if I right-click a Flash object I see nothing more than the audio device and mic volume in Settings… in Vista you have to specify the source explicitly?
    About the Midi player I suggest you VanBasco Karaoke Player, it is good for the Key shift but also to turn off the usual annoying melody track you’ll find in most of midi/karaoke files (I have a post on this topic in Italian, I’ll translate it).
    I’m glad to hear about your duet with Lu! Of course I will assist you ;)

  3. Howard Says:

    Sounds like a good plan, Heracleum, about adding details as needed on separate pages.

    I didn’t make myself clear about the difference between XP and Vista. Both operating systems see the same “Settings” panel in the Flash recorder. But, if I am not mistaken, one must also set the computer’s “default recording device” to “microphone” or to “stereo mix/output” to match the Flash setting. (Am I wrong about that?) It is there that XP and Vista are rather different.

    I should be practicing my Portuguese lyrics, haha! See you at Midomi.

  4. Heracleum Says:

    Ah yes, all right, different Flash settings panel was puzzling me eheh.
    First I will provide configuration details taking winXP as example (since I guess it’s still widely used) then I’ll try to gather info about doing the same on Vista :P
    I you have this one as operating system I’ll bother you later eheh

  5. Lusimar Says:

    Hey guys, I decided to visit this page and what a surprise!You both talking about me!:) Well, I liked it very much Heracleum but I think I’m only able to be the partner, not the mixer as Howard told you!Hmmmm…I don’t know,I need to learn more about that.You understand more about these things than me!!I think you did a good thing posting this “How-To” procedure here.Thank you!!I hope you can do more for us.Did you see that Howard and I will try a duet too?I hope our duet can be so good as your and Fallingsta’s one!Kisses and see you!;)

  6. Heracleum Says:

    Yes Lu, there’s a lot more stuff to add, I’ll also try to post an article about the basic steps a mixer should follow, actually not that easy.. but I can try, maybe a simple example involving just an mp3 karaoke track and not Midi mixing, because it’s quite more complex ;)

  7. Lusimar Says:

    Ah, yes.It would be good because I always use MP3 karaoke tracks on my recordings and I’d love to know how to mix.Thanks for you effort my friend!!By the way, thanks for singing with me!!Kisses!!! ;)

  8. Howard Says:

    Hi Heracleum, Hi Lusi,
    Uh-oh! Lusi says she won’t be the mixer?? We’re in trouble now! Haha!

    But maybe it solves one on the problems I came to ask you about, Heracleum.

    I came here with two topics to discuss: First, I was wrong when I said that the computer’s “default recording device” must match the Flash Settings Panel’s recording device. I purposely experimented by setting my computer wrong, and the Flash correctly. Flash seems to override. So no worries there. Never mind what I said above.

    The second topic is this: In the duet Lusi and I are planning, it starts out with the female voice singing two verses before the male voice comes in. If I were to send her my vocal, how on earth would she be able to start singing her part and have it match up with mine when my part came in?!? It seems impossible! And yet, YOU have managed this same feat in your duet with Lusi, “We’ve got Tonight”. How did you do it?? What’s the secret? (((But if Lusi says she will be the Partner and I will be the Mixer, this might be a moot point for us.))) Still, it’s an interesting question.

  9. Heracleum Says:

    Howard: nice question, I was also sure I mentioned about it in the procedure but I probably left it out to keep the steps clear and short. It depends on how the Partner (P.) -obviously- recorded his/her track: if P. send you a “trimmed” recording, starting immediately with his/her voice only, the Mixer should use his ability (and super-powers :D) to find the point where this recording starts from: it’s not recommended to use “ears” eheh you should use your eyes to graphically recognise the shape of the waveform where the very first word is sang and place it in the correct point. This is why a Midi background music is perfect for mixing: with the help of a melodic track (commonly used in midifiles) you could look at its starting note and vertically align the vocal track with this note.
    It depends on the software and it’s not easy exactly as you were afraid of. Thats’a why the P. should avoid this pain to the Mixer (M.):
    when P. receives the background mp3 track s/he should place it as a first track of his/her multitrack audio software, then records a vocal track. This track of course will have blank spaces (silence) before and after. When P. exports this track the trick (trick-track??) is to create a full mp3 starting exactly where the background track starts, no matter of silence: this way the “origin” (0:00) is the same for both. So when M. receives this vocal track s/he just have to place it at the beginning of the music project, starting at 0:00 as other tracks does. This avoids the hard work to find exactly where the Partner’s vocal track starts.
    Following this directive, if P. has to sing 3 parts in the song s/he should not compact the three parts sticking ‘em all one to the other, because this will multiply the mixing troubles. I hope I’ve explained it well.
    PS: Lu’s track was exactly the worst case possible I mentioned above ahahahah! :D ..but of course it’s all my fault! I forgot (or better, I was too lazy) to mention these steps and I’ve been punished ahah, noo I’m quite used to do such things so no problem at all, but I suppose it’s very hard for a newbie Mixer.

  10. Howard Says:

    I actually think I understood most of that, and it’s bad news! Because I do not have any software such as you describe where I can see the waveform, nor any way of producing separate tracks for a multi-track recording. I used to have a two-track wav file editor, with which I could merge 2 audio tracks into one track, by visually lining up the downbeats. Then with the click of a button, it would create a file that was made of both of the original tracks. It may have been similar to what you described above. That was on an old computer…

    I suppose I will just have to get Lusi to sing her version first, as a solo. Then I will use her recording as my “backing track”. It won’t be as hi-fidelity as what you can do, but it’s the only idea I can come up with. (Unless you have a better idea for a newbie mixer with no software.)

  11. Lusimar Says:

    Uh-Oh!!!!There you are talking about me again! :P So, Howard is right because I don’t know how to mix and I suppose we will need your help Heracleum!! :D Could you help us? I think we (Howard and I) Could send our vocal tracks to you and you could mix them.Would you do this for us, your friends??? ;) Ehh?? Well, I’m sure Howard would be very happy with it like me too!! Kisses for you two my dear and favorite friends! :) See you!!

  12. Howard Says:

    Good news.
    We do need your advice, Heracleum, but we might not need you to do it for us. Not yet, anyway.

    I downloaded a free audio editor today. It’s an updated version of the same one I used to use. It’s called “Sound Engine”. It’s just a 2-track editor, but it’s all digital, so I shouldn’t lose TOO much quality by re-mixing on previous mixings.

    I’ll see what I can do, and maybe ask for more help later. ( (Lusi, send me your test track in the same key as you sang it before. lol) )

  13. Heracleum Says:

    Ok Howard! Yes 2 tracks are good enough :)
    You know, of course for me it’s a pleasure to mix for you everything you want by the way I guess it would be interesting and useful at least to try mixing a simple two track duet.
    Let’s say Audacity is a good choice as a multitrack audio software, also to add effects (usually reverb) to your vocals, so when you (even you both) are ready to “take the long jump” and switch to a semi-professional tool I’d suggest Audacity.

  14. Yayi Dewi Says:

    I’m planning to do a duet with my brother who lives too far from me. Audacity is not too good, so i’m going to do it with Sound Engine.. Let’s keep our fingers crossed and let’s see if a year is enough to prepare our duet. Thanks, le mon!

  15. Heracleum Says:

    Hey Dewi,
    eheh well, a one-year-long duet should necessarily turn out to be a masterpiece!! :)

Leave a reply

Allowed XHTML tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

It sounds like SK2 has recently been updated on this blog. But not fully configured. You MUST visit Spam Karma's admin page at least once before letting it filter your comments (chaos may ensue otherwise).