Tutorial Converting Stereo tracks to Dolby Surround 5.1 tracks. 1. Introduction. In this tutorial, I will explain how to convert a normal stereo (2 channel) sound track to a 6 channels DTS or Dolby Surround track and burn them on DVD that can be played on any home of PC DVD player. DTS and Dolby Digital 5.1 audio is composed of these 6 tracks : 1. FL = Front left and goes to the left speaker 2. FR = Front right and goes to the left speaker 3. C = Center and goes to the central speaker (mainly for vocals) 4. LFE = Low Frequency Effect => deep bass sub-woofer speaker : For very low frequency audio (20 Hz to 100 Hz) 5. SL = Surround Left => Rear left speaker 6. SR = Surround Right => Rear right speaker
Here is how we will obtain these tracks : We will have to create 6 audio tracks from the original 2 tracks. Front-Left => This will be taken from the left channel of the stereo file. Front-Right => This will be taken from the right channel of the stereo file. Center = > We will bandpass filter the stereo file to extract audio in the vocal frequency range (150 Hz to 3 Khz). LFE => We will low-pass filter the stereo file to extract the bass for the LFE. Warning: Extreme care must be taken here to let only signal above 20 Hz in this file. Or else signal below (lower than) 20 Hz may damage your ears or surrounding equipements when it gets amplified by the sub-woofer. Audio editors like Adobe or Wavelab have this control implemented by default but I don't know for other software. SL => Extracted from the stereo left channel, we will add reverbs and flanger effect to make a swirling sound fly on your back. SR => Same as above but extracted from the right channel of the stereo file.
Software I am going to use : – Adobe Audition Version 1.5 or 2.0. Any aother audio editor having the functions mentions in this tutorial can be used. This technique can also be transposed to any other software. – Minnetonka SurCode DVD pro (Or any other freeware that can encode DTS.
Step-1 : Get the audio track and convert to wave (PCM). You can do this with any usual CD ripping software and save the tracks in wav PCM 16 bits format. The resulting file is about 10 Mbytes for every minute of sound. You can also use mp3 files but I advise you to only use High Bitrates (upto 320 Kbits/s VBR). Save this file as test.wav. This file has a sampling rate of 44.1 Khz (The CD specification) but DVD accept only 48 Khz. You can keep the 44.1 Khz if you are only going to burn DTS audio-CD. Not for DVD. We will have to resample to 48 Khz. Open the « test.wav » file in Adobe Audition (From now on I will call it Adobe), Click Edit in the menu bar and select « Convert Sample type »
In the next window select 48000 as sample rate / Stereo as Channels / and 16 as bit depth. Set quality to high by moving the slider to the left till you read 999 then click « ok ». From now on we will use the audio files bundle with this tutorial. These are from the movie : Phir Bhi Dil Hai Hindustani – Song – I'm the best. I have chosen this one as it is in stereo and is a good song. (And finally you will be the best after completing this tutorial). Files included : I'm the best_stereo.flac foobar2000.rar Adobe Audition documentation can be freely downloaded on adobe web site. I have compressed it in lossless FLAC audio format. Use any FLAC decompressor or if you are like me, use FOOBAR2000 (Bundled with this tutorial). You will only have to decompress the original track « I'm the best_stereo.flac ». The file is already in 48000 Hz, so you wont have to resample it. Here is how to decompress in foobar2000 (or fb2k). –
Open the file in fb2k.
Choose WAV (PCM, fixed-point). « ok » and it will be decompressed in the source folder with a « wav » extention.
Now open the « I'm the best_stereo.wav » file in Adobe. Double click on the name of the file in the list and you will see the waveform in the main window. Be sure to click the « edit » button on the left hand corner of the main window.
Now, type « CTRL + L » to select the left channel. Then right click on the waveform and select « Copy to New ». This will extract and generate the left channel as a new waveform. Click on the file name « I'm the best_stereo (2)* » in the left window and save it as « I'm the best_L.wav ». Always save in wav 16 bits / 48000 Hz pcm format. (The default). « CTRL + R » short-cut does the same thing with the right channel. So, do it with the Right channel also and save it as « I'm the best_R.wav ». You will now have 2 new files. Save a copy of them in a seperate folder by naming them. « I'm the best_L_final.wav » and « I'm the best_R_final.wav ». These will be the final front channels tracks.
Show's new waveform created and name is appended with an asterix (*).
Dealing with the surround channels : Now that we have our front channels, lets creat the surround left/right ones : Close all files in Adobe and reload the original stereo file. ( I'm the best_stereo.wav) In the menu bar, click on « effects » then on « Delay Effects » and « Flanger ».
Click on the « play » button to preview what this effect does.
You can see (hear), what it does: Swings the sound from R to L and modulate the loudness. Now reduce the « mix » to arround 30.
You can play with the different options and other effect to sweet your needs. Most commercial movie converted to Dolby 5.1 use the classic «Reverb » effect. For this tutorial we will continue with the « flanger ». Click on the « play » button again to stop the preview. Click « ok » to apply it to our file. Save the file as « I'm the best stereo_flanger.wav ». With « CTRL + L » and « CTRL + R » as previously, copy to new track and save each track to a new file. Name them as follows: I'm the best _SL.wav => This will be the surround-left channel. I'm the best _SR.wav => This will be the surround-right channel.
Creating the Central channel : Ok. Let's creat the central channel. Normally in movies, this hold mainly the vocals. So we will extract the vocal from the original file. Close all files in Adobe and re-open the original stereo file. In the menu bar, click on « effects » then on « Filters » and « Central Channel Extractor ».
For those who knows about devocalising software, this is what the filter do. It attenuates most of the sound components not in the vocal region that are in both tracks by doing an inverse followed by a difference function.
As the main vocal is female we will use the « female voice » preset : In the frequency range tab select « female voice » Click on the preview button to hear what it does. It mostly attenuate lower and higher frequency ranges and emphasizes the vocal frequency (arround 250 Hz to 3 Khz). For male vocal it will be around the range of 150 Hz to 2 Khz. All this can be finely tuned to get best results. The only tuning we will do here is to decrease in the discrimination settings the « crossover » to 0, thus making the vocal clearer (Huh ! And being less « macho » to the lady). Click « ok » to process the file. Then with « CTRL +L » and « CTRL + R » select and create 2 new tracks as before. Save them as I'm the best_CL.wav I'm the best_CR.wav
Now click on the «MultiTrack » button just below the menu bar.
The multi-track window will open, displaying track 1, track 2, track 3 etc... By default it displays 6 tracks but as many as 1024 tracks can be inserted. Right-click on track 1 in the dark region or « track region ». Select « insert » and « « I'm the best_CL ». Beware of the yellow marker, it must be at the very begining of the track (Offset = 0). It marks the begining of the insertion place on the time-line. Insert the « I'm the best_CR » in track-2.
Using the « control » key select both tracks. (They should be highlighted when selected). Then right click in the waveform window. Select « Bounce to new track », then « all audio clips in session mono ». This can also be done with the ctrl+G short-cut. This will mix the two tracks in one mono-phonic track which will be named by default « mixdown ».
The different controls in the left of the multitrack window is not used here. Read through the documentation of the software to see all the options. The main item I will highlight here is that when you downmix tracks (2 or more tracks) care should be taken to the volume level. Here our 2 tracks have level at arround -6 dB, the final mix will result in a -3dB signal (twice louder than the original). This is OK. You should note that in Digital processing, 0 dB is equal to infinite level. So if you mix 2 tracks with -3 dB you will surely have a 0 dB resultant mix. That's not good because the signal will start clipping (very annoying sound effect).
When processing finishes a new track « mixdown » is shown in the left window. Use the « effect » , « amplitude » « amplify » function to raise the volume to arround -2 dB then save it as I'm the best_C.wav This will be our Central track.
Creating the LFE track Close all files and re-open the original track. From the menu click « Effects », « Filters » and « Parametric Equalizer ». Use the settings as bellow. The settings preset mainly : – Fixe the frequency band of the low-pass filter from 20 Hz to 100 Hz. – I use only the second-order filter. – The gain in this band is set to 0. The result is a wave containing only the deep-bass frequency.
Click on the « play » to preview what it does and then click « ok » to process. Use Ctrl+G to copy to new track and the save the file to « I'm the best_LFE.wav ». This is our final track for the LFE section. We donot have to process both tracks cause they contain the same informations.
Preview all the tracks Now lets see what all this will give before converting to DTS or AC3.
Open all the six wave files we have created in Adobe and insert them as follows in the multi-track window : Track 1 = I'm the best_FL.wav Track 2 = I'm the best_FR.wav Track 3 = I'm the best_C.wav Track 4 = I'm the best_LFO.wav Track 5 = I'm the best_SL.wav Track 6 = I'm the best_SR.wav
In the track panel : click on the « Master » button, then select « mono ».
Asign tthe tracks to the appropriate channel of your sound card. Note that this will only work if you have a multi-channel mixer software for your sound card (Creative Audigy2 and many others have). I use an external M-Audio sound module on FireWire and my first 6 channels are set for DTS and Dolby 5.1:
And the settings in Adobe :
Mixer view in Adobe:
In the « mixer » set all levels to 0 dB. Finally select all tracks by ctrl + click then play. And adjust the individual lever to get arround -2 dB to – 0.2 dB. See below.
We can see that : Track 1, 2 & 4 can be boosted by +6 dB. You can do this with « effects », « Amplify » function to the individual tracks, one by one. I personally use Waves MultiMaximizer Plugins to do this in a far better way. (Use of this plugin is explained in my « Digitally cleaning noisy tracks » tutorial. For the next and final step we will not use Adobe so you can close it.
Converting to DTS or AC3. Launch SurCode DVD Pro
Give a name to the destination file : Ex test.dts Insert the corresponding track files and click encode. You will get a DTS file that can be directly burnt to a CD and played in a DTS compatible DVD player. If you want to burn on DVD, use Nero 7 or any other DVD authoring software and insert the DTS tracks. The DTS track will be recognized as an LPCM lossless compression track and can be converted to AC3 if you are making a Dolby Digital 5.1 audio DVD. I'll personally prefer the DTS type of DVD cause DTS has superiour audiophile uncompressed sound far better than AC3 but takes more disk space. Kindly for all music lovers (And good seeders) and specially for the Desi community. MikeKhan (c) CopyFREE May 2006
