This filter resolves the input audio signal into harmonic tones of different frequencies. It allows you to suppress frequencies lower than a specified frequency (so-called high-pass filter) or higher than a specified frequency (low-pass filter). The output audio is the superposition of the remaining frequencies.

Note that an audio signal sampled with frequency *samplerate*
can contain only frequencies from interval [0, *samplerate*/2].

This operation might be useful e.g. to reduce the noise present in audio signal - it is a common wisdom that this can be achieved by low-pass filtering with a suitable frequency.

To perform the high-pass filtering, check the *Minimum
frequency* option and enter the frequency in Hz. To perform the
low-pass filtering, check the *Maximum frequency*
option and enter the frequency in Hz. It is possible to do high-pass and
low-pass filtering at the same time (check both options).

The following table lists the filter's parameters in network file format.

**Table 4.62. BandPass parameters**

Parameter | Values | Description | Required |
---|---|---|---|

audiofilter | BandPass | filter name | yes |

min_freq | nonnegative integer | high pass frequency | no |

max_freq | nonnegative integer | low pass frequency | no |

This plugin uses the FFTW library to compute the fast Fourier
transform of audio blocks. The result of FT is an array of coefficients
corresponding to different frequencies. The coefficients outside the range
`[min_freq, max_freq]`

are replaced with zero and the
inverse FFT is calculated.

This filter can change the number of audio channels from mono to stereo and vice versa. In the former case it just copies the original channel, the latter case averages the two channels in a single one.

Simulates the echo effect when the decayed reflected sound interferes with the original signal.The reflected sound can be reverbed, i.e., it reflects, decays and mixes again and again. "Surround effect" is be created when the echo from the left channel is mixed with the right channel, and vice versa. Also, in stereo a different sound can be obtained if the reflected wave is inverted.

The dialog allows you to choose the channel(s) from which the sound will be reflected, set the length of delay (in seconds), i.e., the time it takes the sound to reflect and return back, the amount of decay (percantage of original signal), i.e., the ratio between the amplitudes of the reflected sound and the original one. It's also possible to turn on the reverb, surround and invert effects.

The following table lists the transition's parameters in network file format.

**Table 4.64. Fade parameters**

Parameter | Values | Description | Required | Default value |
---|---|---|---|---|

audiofilter | Echo | filter name | yes | |

channels | "left" or "right" or "all" | source channels of the sound that is reflected | no | "all" |

delay | real (0.001, 10) | time between the original signal and the reflected | no | 0.1 |

decay | real (0, 100) | ratio of amplitudes of reflected signal to original signal | no | 0.333 (a third of original) |

reverb | boolean (0 or 1) | reverb the echoed signal | no | 0 (reflect just once) |

surround | boolean (0 or 1) | swop the channels in reflected signal | no | 0 |

invert | boolean (0 or 1) | invert values in reflected signal | no | 0 |

This filter resolves the input audio signal into harmonic tones of
different frequencies. One can amplify or weaken different frequencies by
specifying a table `E[f]`

(called the equalizer function),
which means that frequency `f`

has to be multiplied by a
factor `E[f]`

. We are usually interested in continuous
equalizer functions - in this case, it is reasonable to specify only a few
function values `E[f]`

and interpolate the other; this
filter uses linear interpolation.

Note that an audio signal sampled with frequency *samplerate*
can contain only frequencies from interval [0, *samplerate*/2].

The dialog displays a sequence of sliders which represent particular
frequencies - they appear in increasing order and cover the range which a
human ear can perceive. The initial setting is the identity equalizer
function, i.e. `E[f]=1`

. However, the dialog uses the
decibel scale and therefore `E[f]=1`

corresponds to 0 dB
amplification. Move the slider up or down to increase or decrease the
value `E[f]`

. You can always switch back to the identity
equalizer function by clicking the *Set defaults*
button.

The following table lists the filter's parameters in network file format.

**Table 4.65. Equalizer parameters**

Parameter | Values | Description | Required |
---|---|---|---|

audiofilter | Equalizer | filter name | yes |

points | positive integer | number of equalizer control points | yes |

function | string of whitespace-separated integers | list of equalizer control points, exactly 2*points integers | yes |

The *function* parameter allows you to specify
more general equalizer functions than the GUI dialog. It is a string in
format `f_1 E[f_1] ... f_n E[f_n]`

, where
`f_1 < ... < f_n`

and `n`

is the
number of equalizer control points (equal to the parameter points).

This plugin uses the FFTW library to compute the fast Fourier transform of audio blocks. The result of FT is an array of coefficients corresponding to different frequencies. Each coefficient is multiplied by a factor which is determined by interpolating the equalizer function values. The inverse FFT is then calculated.

Normalization is an operation which scales the input samples in such a way that they cover the whole 16-bit range (note that OpenVIP always produces 16-bit audio).

This filter is a generalization of normalization - it enables you to specify the percentage of the 16-bit range the audio should cover.

It is possible to select the channel which should be normalized in
case of stereo sound. One usually wants to normalize both channels (select
*All*); however, it is possible to normalize only left
or right channel, too. In addition there is the *Independent*
option which has the following meaning: *Independent*
toggled off finds the common maximal amplitude for both left and right
channel and then scales both channels with the same factor.
*Independent* toggled on normalizes left and right
channels separately.

The *Volume* slider refers to the percentage of
the 16-bit range that the output should occupy.

In case you switch the *Dynamic* option on, the
normalization effect will vary from the initial setting (*beginning*)
to the final setting (*end*). The last slider affects
the speed of interpolation from the inital to the final setting.

The following table lists the filter's parameters in network file format.

**Table 4.66. Normalize parameters**

Parameter | Values | Description | Required | Default values |
---|---|---|---|---|

audiofilter | Normalize | filter name | yes | yes |

channels | left, right, all | channel selection | no | all |

separate | boolean (0 or 1) | normalize channels independently | no | 0 |

factor | real from [0,1] | volume factor (initial setting) | no | 1 |

factor_2 | real from [0,1] | volume factor (final setting) | no | factor |

rate | positive real | interpolation rate | no | 1 |

This audio filter is capable of shifting stereo audio either to left or right channel.

The *Balance* slider specifies the amount of
shift: 0 means no change, -100 mixes the right channel with the left (and
leaves the right channel silent), 100 moves the sound completely to right
channel. The other values just interpolate between the two extremes.

The *Dynamic* switch allows you to start with an
initial shift setting and proceed to a final setting; the slider
*Rate* then sets the speed of interpolation from
initial to final setting.

The following table lists the filter's parameters in network file format.

**Table 4.67. PanChannels parameters**

Parameter | Values | Description | Required | Default values |
---|---|---|---|---|

audiofilter | PanChannels | filter name | yes | yes |

shift | real from [-1, 1] | channel selection | no | all |

shift_2 | real from [-1, 1] | normalize channels independently | no | shift |

rate | positive real | interpolation rate (1=linear) | no | 1 |

Changes the sampling frequency of input audio.

The dialog allows to choose new sampling frequency (in Hz) and the type of interpolation used during resampling (nearest neighbor is quick, while linear interpolation reduces the danger of alias effects).

The following table lists the transition's parameters in network file format.

**Table 4.68. Resample parameters**

Parameter | Values | Description | Required | Default value |
---|---|---|---|---|

audiofilter | Resample | filter name | yes | |

freq | integer > 1000 | new sampling frequency | no | 0 (retain the frequency) |

interpolation | "NN" or "linear" | "nearest neighbor" or "linear" interpolation of samples | no | "linear" |

This filter mutes the input audio.

This filter simply exchanges left and right channel of a stereo sound. It makes no change to mono audio.

Changes the volume of audio in selected channels. The changes may be dynamic and the rate of changes can be set.

The dialog allows to choose the channel(s) that will be affected as
well as the factor multiplying samples of the original audio sound. If
'dynamic' box is checked, it is possible to set also the factor
for the end of the audio. The actual factor is computed as an
interpolation between these two and the rate of interpolation can be set.
If factor is negative, the audio samples are inverted (e.g.,
`factor = -1`

reflects the wave about the time axis).

The following table lists the transition's parameters in network file format.

**Table 4.71. Volume change parameters**

Parameter | Values | Description | Required | Default value |
---|---|---|---|---|

audiofilter | VolumeChange | filter name | yes | |

channels | "left" or "right" or "all" | channels affected by volume change | no | "all" |

factor | real | the factor the audio values are multiplied by | no | 1.0 (retain volume) |

factor_2 | real | the factor the audio values are multiplied by at the end | no | factor (retain volume) |

rate | real > 0 | the rate of interpolation of factor | no | 1 (linear interpolation) |