Instruction manual - Pizzicato 3.3 EN900 - Revision of 2008/04/30

Light

Beginner

Professional

Back to the Pizzicato main site

Audio tracks

Subjects covered:

Audio or MIDI? [Light] [Beginner] [Professional]

Up to now, we have only used MIDI to manipulate music and hear the notes. Pizzicato plays the score by creating a series of MIDI events. The sound card of your computer or your external synthesizer uses those MIDI events and transforms them into a real sound that you can hear through the speakers. This working method is quite satisfactory in many cases. Most music information can be indeed represented by a sequence of MIDI events.

However, there are cases where MIDI can not be used. For instance, if you want to add your voice to a music arrangement, or if you want to play a real instrument and record the exact way it sounds, MIDI is of no use. In that case, the program must record and store the sound itself (or at least its mathematical representation).

As we have seen in a previous lesson (see the "Music notation" lesson), a sound is simply a vibration of air. It has four main qualities: the pitch, the amplitude, the timber and the duration. When an acoustic instrument plays, it vibrates at certain frequencies and transmits that vibration to the air around it. This vibration finally comes to your ear and you get the sensation of sound, with its various qualities.

To record sound, we need to measure the position of a particle of air, around its central position, so that we can thereafter reproduce the same vibration. For practical purpose, we can say that the zero value will represent the position of a particle of air when no sound travels through it. If the particle is pushed forward, we can measure its displacement from its central position and write it down. Such a measure is called a sample.

As the vibrations of sound are quite fast, we must do so a lot of times per second. In fact, as the human ear can only perceive up to about 20 to 22 thousands vibrations per second, there is a mathematical principle that tells us that we must measure the air position at least two times that frequency, so that the sound is recorded and stored accurately. This is the reason why, on a music CD, the frequency has been fixed to 44100 samples per second. This frequency is also called the sample rate and is expressed in Hertz (abbreviation Hz) which means "vibrations per second".

If we record a sound with a much lower sample rate, a part of its vibrations are lost when we reproduce it. Indeed, the information is missing in the record and the sound quality deteriorates. The most common example is a telephone line. In the beginning, they used an 8000 sample rate. As the human voice is not using the full spectrum of sound vibrations, the phone quality was enough to transmit an understandable conversation. But try to listen to a full symphony orchestra through a phone and you will hear that the sound quality is quite poor, indeed.

Another important parameter is the precision with which the position of an air particle is measured. You could measure on a scale that contains 10 positions or 10000 positions or more. If there are too few positions on the scale, there will be an approximation of the position. To give a comparison example, if you measure the dimensions of a room with a meter that do not have any centimeter graduations, you could then say the room is 3 meters or 4 meters, but you could not say exactly 3 meters and 23 centimeters. By transmitting the information that the room is 3 meters, it would not be exact and if an architect creating a big building would use that precision level, there is no doubt that the building would fall apart very soon. So the precision is quite important. For the CD quality, a scale has been set that goes from -32768 to +32767. The specificities of those numbers come from the binary system used to encode them. Indeed, 2 to the 16th power equal 65536 so that there are so many possible values to a sample (65536 = 32768 negative values + 32767 positive values + 1 value for zero). So we say that the quality of a CD is in 16 bits (a bit is a BInary digiT).

As a summary, we can say that an audio recording is a series of numbers that represent a sound and that may be used to recreate that sound at any moment through the speakers of a computer. In CD quality, one second of music is represented by 44100 times 2 numbers (because stereo is used, so there are two channels), each one of them being in the range from -32768 to +32767. When we put those numbers on a graph with time, we get a picture like this:

where the upper part is the left channel and the lower part is the right channel. Those curves represent in fact the position of an air particle when this sound is heard. Most software working with audio will represent sound as above.

 

What is an audio track? [Light] [Beginner] [Professional]

An audio track is the representation of a sound for a specific duration. For instance, you could have a track of 4 seconds in duration. By playing this track, you will hear the sounds it contains.

Pizzicato lets you add one audio track (Beginner and Light versions) or several audio tracks (Professional version) to a score. You can specify when the audio track begins, in relation to the score and Pizzicato will play it together with your score.

Each audio track is in fact represented by a wave file (".wav" extension on Windows). There are lots of existing audio file formats. Pizzicato works with WAVE files, as they are very common and can be used on Windows as well as on the Mac. You will find various software format converters on the Internet, so that you can use wave files and then convert them into or from mp3, aiff and many other audio formats.

You should well understand that the audio information is stored in the wave file and not in the Pizzicato document. The Pizzicato document contains a reference to the wave file, so that it can play it at the same time than the score. If you want to send a Pizzicato score containing audio tracks to a friend, do not forget to send also the corresponding audio files. Existing audio files may be inserted in the score, but you can also create an empty audio file and record it yourself, with a microphone.

Adding an audio track [Light] [Beginner] [Professional]

Depending on the Pizzicato version you have, you access or add an audio track differently. For Pizzicato Light and Beginner, do the following:

For Pizzicato Professional, do the following:

The audio track selection dialog appears as follows:

This dialog lets you specify the associated audio wave file. You may either select an existing audio file or create a new one. We will first open an existing audio file.

It is an 8 bits, mono, audio file, with a sample rate of 22050 samples per second and its duration is 6 seconds and 122 milliseconds.

You may also create a new audio file and associate it with the score.

The Delete button is used mainly for the Pizzicato Light and Beginner versions. Clicking on it just removes the reference to the audio file from the Pizzicato score. Notice that the audio file itself is not destroyed.

Multiple audio tracks - Pizzicato Professional [Professional]

Pizzicato Professional only:

With Pizzicato Professional, you may add several audio tracks to the same score. You may start the audio tracks at different position in the score. When an audio file has been added, it appears just below the score as a colored band, at the position where it will start to play and with the length showing its duration in relation to the score:

The little icon on the left bottom of the score rectangle shows that this score has associated audio tracks. Clicking on it switches between viewing the audio tracks below the score or not viewing them.

To add another track, just do exactly the same as explained above. In fact, you may add the same file more than once, at various positions in the score. When adding more tracks, the conductor view will display something like this:

For each track, you may specify the starting point in the score as well as the volume of the audio file. The starting point may also be adjusted by directly dragging the yellow bands horizontally. While doing so, there is a magnetic grid that positions the starting point at multiples of half a beat, so as to accurately place the audio file, but you may disable that magnetism by holding down the CTRL key.

When you click with the right mouse button (alt-click on Mac) on an audio track, three menu items appear:

Audio setup [Light] [Beginner] [Professional]

The audio setup should work correctly by default. You may access it in the Options menu, Audio setup...:

The Default values button lets you reset the parameters to their default values.

If you experience a delay between the score (MIDI instruments) and the audio track, you must add the same delay to MIDI, so as to compensate. This is done by the MIDI play options... item of the Options menu. There is a slider named MIDI delay in milliseconds (correction for audio). Adjust its value so that MIDI and audio are played exactly togther.

Back to the Pizzicato main site

Light

Beginner

Professional