About Sound Studio Filters and Effects.

Filter Menu

A filter applies an effect on the selected audio. An effect works by taking the audio as input, applying some kind of transformation to it, and putting it back in the file. If nothing is selected, the entire file will be selected before applying the filter.

Some filters will let you save presets. A preset allows you to save the position of the controls of an options window so that you can recall them later. You can save several presets in the default presets folders. Any files in the default presets folder will show up in the “Presets” popup. To create a preset, select “Create New Preset” from the pop-up button, give it a name and save it. Then make changes to the options window. Any changes will be immediately saved to the presets. To make changes without affecting a preset, select “Default” from the presets pop-up.

To rename, delete, or otherwise manage presets, select the “Reveal in Finder” command from the pop-up. This will show the presets folder in the Finder, where you can manage the preset files. Presets are stored in the “~/Library/Application Support/Sound Studio/Presets/” folder, in your home directory.

Last Filter

The first item in the menu shows the name of the last filter used. Selecting it will reapply the filter to the current selection without showing its options window.

Amplify/Volume

This command will raise or lower the volume of the selected audio. The options dialog box shows the amount of the change in both decibels (dB) and percentage factor (%). To adjust the volume to a uniform level, see Dynamics Compressor and Normalize, below.

Fade In

This command applies a linear fade from zero volume to full volume.

Fade Out

This command applies a linear fade from full volume to zero volume.

Fade Special

This command allows you to draw and apply a fader envelope over the waveform to create custom fade effects. There will always be a start point and an end point, but you can add several points in between. You can drag these points up and down to modify the envelope. To delete a point, hold down the option key and click on the point. The vertical scale is in linear amplitude. The horizontal scale is in time, as a percent of the selected time. The actual time scale will depend on what is selected.

Normalize

This command normalizes the volume of the audio. If loudest peak level is chosen, then this command scans the selected audio for peaks, which are the tops of the waveform, and adjusts the volume so that the peaks coincide with the target level specified in the options dialog box. If loudest RMS level is chosen, this command adjusts the volume based on the average power of the waveform, which closely follows what our ears perceive to be the level of the audio. The target level is shown in both decibels (dB) and percentage factor (%), with 0 dB being the maximum bandwidth of the file, and lower levels being negative decibel numbers.

Normalize each track independently: This option will treat each track separately when searching for and adjusting the peak levels. When multiple tracks are selected, this option will effectively bring each track to about the same volume relative to each other.

Normalize all tracks together: This option will treat all track as a unit when searching for and adjusting the peak levels. When multiple tracks are selected, only the highest peak level of all tracks is considered. Each track is adjusted by the same amount so that the volume of each track relative to each other will stay the same.

Dynamics Compressor

A compressor reduces differences in volume between quieter and louder sections of audio. It is applied to audio which goes above the threshold level by turning the volume down by an amount determined by the compression ratio. The attack time is how fast it responds to levels above the threshold, and the release time is how fast it reacts when the levels drop below the threshold. The post gain increases the overall volume to compensate for the lowered volume in the formerly loud sections.

Another way to describe the dynamics compressor is to imagine someone watching the peak level meter while the audio is playing, and whenever the level goes above a certain threshold, he turns down the volume. This is what a compressor does. The compression ratio determines how far to turn down the volume when the level goes above the threshold, with larger ratios resulting in the volume being turned down more. The attack time determines how fast the compressor reacts to the level going above the threshold. The release time determines how long after the level goes below the threshold the volume is returned to normal.

The “Post Gain” control compensates for how the compressor turns down the volume on the loudest passages. It does this by amplifying the audio after it goes through the compressor. Turning on the “Auto” option causes it to calculate the optimum post-gain, so that the volume of the loudest passages stays about the same but the quieter passages are increased in volume. The effect of this is to bring up the level of quieter passages without increasing the volume of louder passages, essentially compressing the dynamic range of the audio.

Compression is useful for taking music with a wide dynamic range, where the quiet passages are very quiet and the loud passages are very loud, and making all the passages about the same volume so that they can all be heard in a noisy environment such as in a moving car. Compression is also used when someone is speaking into a microphone, so that if that someone moves closer to or farther from the microphone, or speaks softer sometimes and louder sometimes, the compression can compensate for these differences and output a consistent level of audio.

The Peak/RMS options under Threshold determine what kind of level meter to use. A peak level meter measures the very tops of the waveform, and is triggered whenever the wave goes above the threshold. A RMS level meter measures the average power of the waveform, and closely follows what our ears perceive to be the level of the audio.

Dynamics Expander

An expander increases the differences in volume between quieter and louder sections of audio. It works by turning down the volume when the volume level stays below the threshold level, and turning the volume back up when the level goes up above the threshold. The attack time is how fast it responds to levels above the threshold, and the release time is how fast it reacts when the levels drop below the threshold.

Another way to describe the dynamics expander is that it is the reverse of a compressor. An expander works by reducing the volume of the audio when the levels drop below a certain threshold. The ratio determines how much to turn the volume down, with a larger ratio resulting in the volume being turned down more. A high ratio of 12:1 or more is considered a noise gate. The attack and release times behave the same way as in a compressor. The attack time is how fast it reacts to the level dropping below the threshold, and the release time is how long before the volume returns to normal after the level goes above the threshold.

Expanding is useful when you want to increase the dynamic range of the audio. It is also useful if you have a noisy recording and want to reduce the volume of the quieter passages so you don't notice the noise as much. It does have the side effect of changing the way sounds decay and can end up silencing some parts that are quieter.

Noise Gate Expander

The Noise Gate is a version of the Expander which turns the volume all the way down. It generally silences any audio that falls below the threshold level. The attack time is how fast it responds to levels above the threshold, and the release time is how fast it reacts when the levels drop below the threshold.

This is most useful on percussive instruments such as drums and on spoken word recordings, where you have silent gaps between parts with audio signal, and you know you want the gaps to be completely silent.

Add Noise

Adds white noise to the current selection. You would use noise to mask out other sounds in the recording.

DC Offset

This command can be used to shift the waveform vertically in the display. In the manual mode, you would enter the amount to shift the waveform, and the entire selection will move by that amount. In automatic mode, it removes any signal below 20 Hz, resulting in a signal that is centered on the zero voltage line. This can help increase the dynamic range of a recording.

Interpolate

This command can be used to repair individual spikes in waveforms by drawing a smooth line from the start of the selection to the end of the selection. For best results, zoom in to 1:1 and select the area just around such a spike. If you select the entire file and apply this filter, you will get a straight line from the start to the end of the file, which is not how you want to use this effect.

Invert Signal Polarity

Changes the waveform so that positive voltages become negative voltages and vice-versa. In the waveform view, this has the effect of flipping the waveform on the x-axis, so that it is upside- down relative to the original. You won't hear a difference after you've applied the filter, but this filter is useful if you reversed the positive and negative terminals on your cable connections, and you ended up recording the audio with the voltage inverted. Applying this filter will fix the inversion.

If you consider waves that go above the centerline of the display to be positive and waves that go below to be negative, the Invert filter just makes the positive parts negative and vice-versa. This filter is useful when you have a stereo file and one of its channels is inverted relative to the other. The audio will sound like it's coming from the sides when you listen to it in stereo, with no audio coming from the center.

Swap Left and Right Channels

When there are two tracks selected, this command swaps the two tracks. With more than two tracks, it swaps the odd number tracks with the even number tracks.

Backwards / Reverse

This reverses the audio in the file so that it sounds like it is playing backwards.

Chorus

This effect adds the effect of having two voices or instruments playing the same tune, but since they're not perfect, they are slightly out of tune and out of time with each other. It does this by applying a variable-length delay to the audio and mixing the delayed audio with the original audio. An LFO (low-frequency oscillator) controls the amount of delay so that it is oscillates between a longer delay and a shorter delay.

The “Dry Mix” is the original audio, and it adjusts the level of the original audio. The “Wet Mix” is the delayed audio, and it adjusts the level of the audio with the variable-length delay applied to it.

The “Cycle Time” controls how fast the LFO (low-frequency oscillator), which drives the variable- length delay, runs. Changing the speed also changes how far out of pitch and time the “wet” audio becomes. A higher speed causes the LFO to run faster, which means that the “wet” audio is being sped up and slowed down much faster.

The “Minimum Delay” controls how long of a delay the “wet” audio will always have. This prevents the “wet” audio from matching up exactly with the “dry” audio.

The “Sweep Depth” controls how far the delay is allowed to go. Increasing this will also cause the pitch and time to vary more.

The “LFO Waveform Shape” controls how the LFO sweeps across the delay time, either in a gentle sine wave or in a more abrupt triangle wave.

Delay and Echo

This effect takes the original audio as input, delays it, and mixes the delayed version with the original version. You can control how much of the original audio (the “dry” mix) and delayed audio (the “wet” mix) are mixed into the resulting audio. You can also have the delayed audio take its output as its input, resulting in feedback, which is controlled by the “Feedback” checkbox.

Without the feedback, the delay is like having a tape loop which records what you say and plays it back a certain amount of time later. With feedback, it is like shouting across a valley and hearing your voice echo back to you over and over until it dies out.

The “Delay Time” determines how long the delay line is, or how much time there is between each echo.

Note that the effect will end abruptly at the end of your selection, or the end of the file. If you want the delay or echo to decay naturally, you will need to select a few seconds beyond the sound, first adding some silence to the end of the file if necessary.

Flanger

This effect creates a “flanger” sound in the audio. Technically, it is the same as the chorus effect, but with much shorter delay times. The effect causes a sweeping whoosh sound in audio with a lot of broad frequency content, such as hi-hats. You can try the effect by inserting about five seconds of noise and applying the filter with the default settings.

See the Chorus effect for an explanation of the controls.

Pitch and Tempo

This effect allows both pitch and tempo to be changed independently. Pitch is how high or low the notes sound. Tempo is how fast or slow the notes are played.

Pitch is measured in cents, which is one-hundredth of an equal tempered semi-tone. One semitone is 100 cents, and one octave is 1200 cents.

Tempo is measured in percent of the original file's tempo. Fifty percent is half speed, and 200% is double time. Tempo also affects the total duration of the file.

If you want the traditional, analog-style pitch control, turn on the “Link Pitch & Tempo” checkbox. This turns off the pitch shifting compensation and just varies the playback speed.

Reverb

This effect adds a natural reverberation effect to the selected audio. When you are in a room, a hall, an auditorium, a stadium, or any other kind of enclosed chamber, the sounds you hear have some kind of reverberation because of the sound waves bouncing back and forth between the walls, the floor, and the ceiling. This effect is most noticeable in a large enclosed stadium, where the announcer's voice echoes through the stadium. You first hear the announcer's voices, and then you hear several, less distinct echoes of the announcer's voice. Usually you don't notice reverb because your ears are used to hearing it, but without it, the audio sounds flat, dry, and lacking in character. Our ears use reverb to define the size and shape of the room we're in.

Audio signals in the computer are often recorded without any reverb. If you record an instrument directly, or if you use a unidirectional microphone or one close to the sound source, you will get little or no reverb in the signal. To make the audio sound grander, we add reverb.

The "Room Size" controls how long the audio takes to bounce back and forth between the walls. The shape is fixed in this effect, but you can adjust the size, so you can simulate anything from a small room to a large stadium.

The "Decay Length" controls how long the reverberations can be heard bouncing between the walls. A short decay means that the reverberations die away quickly, while a long delay means that they can be heard longer. Generally, a bare room with hard surfaces like tile and stone reflect sound well, and will allow the reverberations to keep bouncing around longer. A room with carpets, drapes, and lots of soft furniture will cause the reverb to die away very quickly because all those soft surfaces absorb the sound.

The "Low-Pass Filter" simulates the effect that sound going thorough the air and bouncing off of softer surfaces will tend to lose their higher frequencies. This option can be adjusted to help color the sound of the reverb.

The "Level" controls how loud the reverbs are compared to the original sound. Typically, you want to keep this low so that the reverb doesn't overwhelm the original audio and make it a mess. You can set the level high if you want the effect of a bad PA system in large stadium where you can't tell what the announcer is saying.

The reverb filter uses six comb filters with low-pass filters and a simple delay. The "Room Size" parameter controls the delay length of each comb filter. The "Decay Length" parameter controls the amplitude of each comb filter. For stereo files, the effect is applied equally to both channels.

A comb filter is a type of filter where the input audio is delayed by a constant amount of time, and then fed back into itself. It is called a comb filter because it will cancel out those frequencies that are 180* out of phase when the delayed audio is fed back upon itself. When you look at the output of such a filter on a frequency graph, it will look like a comb.

3-Band EQ

This effect allows you to change the levels of bass frequency, mid frequency, and treble frequency audio content.

10-Band EQ

You can think of audio as being made up of many different frequencies of sound all mixed together. Your ear has nerves which are sensitive to various frequencies, with the lower frequency nerves detecting a bass guitar while the higher frequency nerves detecting the violins. In reality, these instruments generate many different frequencies of sound at the same time, and the different level of each of these frequencies contributes to the tone, or color, of their distinctive sounds.

Using the sliders, you can adjust the levels of each of these frequencies to color the sound. If the highs sound too harsh or tinny, you can reduce the higher frequencies. If the bass seems to be too weak, you can increase the lower frequencies. If you want to bring out the vocals, you can increase the frequencies around 1 kHz. Ultimately, you will need to use your ears to find the ideal equalization for your audio.

Graphic Equalizer

This filter allows you to draw in a chart to represent how to adjust the levels in each frequency band.

High Pass

This filter will allow only frequencies above a certain frequency to pass, thereby removing lower frequency audio. You can use this filter to remove lower-frequency sounds. The frequency control determines the cut-off frequency, or the point at which the curve hits the -6 dB level, and the “Steepness” control determines how many samples the filter uses as input, indirectly affecting how sharp the drop-off is at the cutoff frequency. You can use the graph in this dialog to see how the filter affects different frequencies.

One effect you can get with this filter is making audio sound as if it were being played on a cheap radio with tinny speakers. Another effect is to use this filter and the Low Pass filter to emulate audio coming through a telephone.

Low Pass

This filter will allow only frequencies below a certain frequency to pass, thereby removing higher frequency audio. This filter is useful for removing high-frequency noise due to artifacts from the digitization process. You can also use if you know your source material doesn't have any audio content above a certain frequency, and you want to remove the noise above that frequency. You can also use this filter to create certain effects. One effect is to make the audio sound like it is being played far away and muffled by walls and other things between you and the sound source.

The frequency control determines the cut-off frequency, or the point at which the curve hits the -6 dB level, and the “Steepness” control determines how many samples the filter uses as input, indirectly affecting how sharp the drop-off is at the cutoff frequency. You can use the graph in this dialog to see how the filter affects different frequencies.