If you've got a load of tracks on vinyl or tape, and you want to MP3 them, it's really dead simple.

Equipment

Firstly, you need a computer (doh!), preferably pretty quick, especially if you're going to rip on the fly.

In that computer you want a soundcard with a Line In socket. Don't be tempted to use the Mic in socket, as it'll probably blow your soundcard.

Next, you want a music system of some description. You can even use a walkman, so long as it has a headphone socket.

The final piece of hardware is a cable available from your local electronics shop (Tandy, or Maplin) which has a headphone-socket connection at both sides. If your hi-fi system has an old style connection, you'll want to get the adaptor to plug into it too. The cable will set you back about £2.50 ($3.60, or 8.80 Guilders), or £3 ($4.30, or 10.50 Guilders)(if you get the gold-plated connections).

You'll also need some software: you can use stuff like Windows Sound Recorder, but you'd be better off with something like Audiograbber (which I used because I had it on a shareware CD).

Setting Up

Stick one end of your new cable into your Line In port, and one end into the music system, after reducing the volume to its lowest level (to preserve your speakers). Check your volume levels - under Windows you can do this by double-clicking the speaker, Options, Properties and selecting Recording properties. Mute everything except for the Line In, which you want at maximum volume. Move back over to the playing settings, and make sure Line In is unmuted. If you've got a power chart, adjust the volume until it's just peaking at the top, and not staying there too long, then record using your chosen program - you may need to rip separately if you use a standalone WAV format.

To increase quality, use good quality hardware: gold cable, good soundcard, good source media and a good hi-fi system.

Here's a detailed guide to converting your tapes or records to mp3s. Though not every step is required (it's actually very long and sometimes a sheer overkill), I've included all tricks I've used just to be comprehensive. I plan to update this as often as possible too. Use the techniques as you see fit.

Some technical stuff regarding audio tapes. Audio tapes are analog stuff. Hence they generally have bad signal-to-noise ratios, dynamic range, fidelity, frequency response etc. when compared to digital audio. Therefore when converting to digital, we cannot expect the same quality as say a CD. The tape can only reproduce signals of frequency 50-16000Hz (to 17000 Hz for Metal tapes) at the most as compared to the 5-20000Hz of a CD. Hence,giving a too high bitrate hoping for a better sound may in fact turn out to be a waste.

Step 1: Setting up the hardware
First of all you will need a good tape deck (preferably a hi-fi rather than a walkman) or record player. Make sure the tape heads and tape transport mechanism are clean. You may have to clean the head more often if you use old tapes. You could also make sure the head alignment is correct, and demagnetize the tape heads (if you know how to do such things). A lot of bad recordings are the result of skipping these simple steps. Now turn the volume to minimum and connect the audio output of the tape deck to the PC sound card Line In socket. The plug at the PC end is like the plug used in those mini headphones and that at the tape deck end is the same if using the headphones output or a standard RCA Audio Jack if you are using the Line Out of the tape deck. Most people prefer the headphones type plug. This cable is easily available or you can create your own with a pair of such plugs and a two-core shielded cable (it has two wires for the Left and Right channels and a metal shield for protection from interference. Lay the cable far from the PC power supply and connect it firmly to both the PC and the tape deck. Check that no 'hum' is heard - it's a bit like the sound you hear from power transformers. If there is a 'hum' check the connections and keep the cable a good deal away from the mains. If you are the electronics type, you would like to build a little pre-amplifier - a must for recordplayers in order to provide amplification and phono equalization as recommended by the RIAA. Many such circuits are available on the Internet. You can include an additional amplifier on the LP source as it lacks a proper powered output (thanks for the suggestion, belgand)

Step 2: Testing the setup
Now, switch on the PC, and open the sound mixer or Windows volume control.('Start>Run>sndvol32.exe' or select it from the start menu). Mute all inputs except Line In. Play a sample tape and adjust the volume on the tape deck to the usual level. Now adjust the level on the Windows volume control till the volume level is comfortable.

Step 3: Choose a recording software
In order to record the sound, you may use any utility, from the basic Windows Sound Recorder to sophisticated sound editors like Cool Edit, Goldwave orSound Forge. These editors are recommended as they allow you to do a lot of 'cleaning up' of the audio and enhancement. I shall explain the method using mainly Cool Edit.

Step 4: Choose Sample Rate etc.
A sample rate of 44100/sec with 16-bit Stereo is usually chosen for recording as it is the same as the CD and all sound cards support it. It gives a good compromise between size and quality.

Step 5: Set recording levels
Play the sample tape and adjust the volume level keeping an eye on the decibel meter (sound level meter) on the editor. (In Cool Edit, this is at the very bottom). The units are in decibels. The level should be high enough so that noise is minimized,but not too high or a form of distortion, called clipping occurs - a form of overload, where the waveform is 'clipped off' at the peak, as if by a pair of scissors. This can be seen as a flattening of the waveform peaks in the sound editor view. You can test the onset of clipping by increasing the volume and checking if the level meters appear as if they are limited at a certain point and they won't go further up. This is the clipping point and while recording, the signal should never exceed this. There is a misconception that the signal can always go up to the 0 dB level (i.e. maximum level in layman's language) on the graph. This is not always true, especially if you have a sub-standard sound card. You may find that the maximum level is say -2 dB or so actually. So be careful. After playing around with the controls, you could come to a good compromise for the recording level.

Step 6: The recording
Play the track on the tape from a period on the tape a few seconds before the track actually starts and start recording. This is to stabilize the tape transport and to record a portion of the background noise on the tape (the 'hiss' that you hear), which will be used to filter out the noise from the track later. Continue recording until the track is over. Save the file. You can also record multiple tracks which you can split later.

Step 7: Processing
Sound Editors offer a lot of ways of improving the perceived quality of sound.So of the features you can use are:

  • Noise filtering: The background 'hiss' of the tape can be removed by first taking a sample of the noise - from the beginning of the recording before the actual music as mentioned earlier, and saving the noise profile. Different sound editors have slightly different methods (Read the manual). Now select the whole recording and apply the Noise Reduction transform to it.Higher noise reduction settings may introduce distortion so keep a careful balance. Cool Edit provides a preview facility so you can dynamically adjust settings. 'Hum' may also be removed by filtering out the precise frequency (mains frequency - 50Hz in most places, 60Hz in the US) and its multiples(100Hz, 150Hz etc. or 120Hz, 180Hz etc.) by means of precise filter transforms like Cool Edit's FFT filter. Settings Presets are provided for convenience so that you don't have to go too deep into the technicals. Select 'Hum removal' from the presets. Another trick is to use the Noise Gate Transform - a kind of on/off switch for the sound, that shuts down output when only noise is present.
  • Clip Restoration: As mentioned earlier, clipping is to be avoided while recording itself. However, if you later find that an existing recording is already clipped, and you have no option of getting a new recording, Clip Restoration (a feature of Cool Edit) can be used to minimise the damage. Roughly speaking, it shapes the flattened clipped areas so that they apparently reduce the distortion of the track. But prevention is always better than cure...
  • Click/Pop Removal: This can be used to remove the 'clicks' or 'popping' noises (like pop corn) from tapes or record tracks. It basically looks through the song for sudden changes in amplitude and interpolates them out, removing these sudden spikes.
  • Editing: You can cut off the leading and trailing silence blocks on the track, or split the recording into several tracks, similar in fashion to a word processor.
  • Enhancement: You can tinker with the Graphics Equalizer Transform (similar to the graphic equalizer on Winamp) to tweak the sound to your own liking, though most people tend to make things worse by using this. The sound may not improve much this way, because it only plays around with the frequencies already in the track and does not recreate anything lost while recording onto the tape in the first place. A better idea would be to artificially recreate higher frequencies that were lost while recording onto the tape. Some editors (like Acoustica for example) have such enhancement transforms and makes the track sound a little better. Though editors offer several effects like echo, flange, distortion etc., it is better not to try such tricks on a neat audio track.
  • Amplitude and Frequency adjustments: Sometimes, due to head misalignment, faulty tapes or loose connections, one channel (left or right) may have higher amplitude or bandwidth as compared to the other. If the problem cannot be corrected, you can artificially increase the amplitude of one channel only to make them equal. For bandwidth problems the effect is more irritating. One of the channels may have a proper sound while the other may have a bass (lacking in treble frequencies). This totally destroys the stereo image of the sound. A (rather complex - the beginner may ignore) trick that can be used is to take a copy of the high frequencies of the 'good' channel and add it to the deficient channel. In Cool Edit Frequency Analysis window, perform a scan on the entire song and obtain the frequency statistics of the track. Comparing the frequencies characteristics of the two and note the region where the 'bad' channel is deficient. Now extract the frequency band from the 'good' channel by means of the FFT Filter and Mix-Paste it into the 'bad' channel. Note that although the track may sound better, it still does not get a good stereo image.
  • Dynamic Compression and Expansion: Dynamic compression makes loud sounds quieter and quiet sounds louder and expansion does the opposite stuff. Worth experimenting with, since tape recordings generally have lower dynamic range than CDs and may also have been processed through dynamically-compressing noise reduction schemes like Dolby B/C
  • Others: Several features are provided by sound editors and not all of them can be covered here. Some tricks that can be done are mono-stereo conversion, wow and flutter reduction, etc. I hope to cover some of these later.
After these operations, the track can be saved to a wave file for final conversion to mp3.

Step 8: Conversion to mp3
Of the many encoders available to convert wave files to mp3 files, the best are FhG (Fraunhofer), available with Cool Edit and LAME, an open-source encoder. The latest FhG encoder has support for the new mp3pro format that sounds very much like regular mp3 but uses less space by using a technique of recreating higher frequencies lost while conversion. Meanwhile, it can also be played back on standard mp3 players though with lesser quality. However, it is better to encode the wave file to standard mp3 using LAME (as it is free and gives good performance). The Xing encoder is very popular (AudioCatalyst) but is not as good as FhG or LAME. Another encoder is the free Blade. A graphical front end for LAME (which is command-line by the way), is RazorLame (again freeware). To encode, add the wave files to the list to be processed and give the proper setting to encode the tracks. Generally 128kbps is sufficient for tape recordings, as the quality of tape recordings limits the quality of sound that can be obtained. Use Joint-Stereo, and Quality Setting 2 or lower. It is better to use the LAME variable bitrate ('adjustable bitrate' in LAME terminology) with average bitrate 128kbps and spread (64kbps-320kbps). Refer to the LAME manual for further details. The command line should be something like '-m j --abr 128 -b 64 -B 320 -q 2'. After conversion, it is a good idea to fill up the ID3 tag, preferably ID3v2 (with Winamp or any mass renaming software like Tag & Rename or Abander) and to rename to file to Artist -Title.mp3 format. Thus you have the final mp3. Phew!

Check out pub41.ezboard.com/br3mix, the r3mix forum for detailed discussion on mp3 and other formats.

That was a rather long explanation, but you need not follow all the steps. (In fact not all are actually used most of the time). They were just inserted there for completeness. And Happy encoding.

Log in or registerto write something here or to contact authors.