Recherche Google
OpenMusic DocumentationOM 6.6 User Manual > Audio > Sound Processing
page précédentepage suivante

Sound Processing Functions

Sommaire
  1. General Principles
  2. Processing Functions
  3. Cascading Processes
  4. Automating Processing

There exist a number of functions allowing to manipulate sound and audio data. A number of them are based on the underlying LibAudioStream library functionalities.

General Principles

The LibAudioStream library used by OM manipulates sound resources under the form of " streams ". Streams can be seen as abstact pointers representing a sound and possible transformations. In OM the LibAudioStream streams are embedded in a structure called SOUND-DATA.

The SOUND-MIX function creates a SOUND-DATA pointer corresponding to the mix of two sounds.
The SOUND-MIX function creates a SOUND-DATA pointer corresponding to the mix of two sounds.

SOUND-DATA pointers can eventually be rendered and converted back to sound file using the SAVE-SOUND function.

SOUND-SAVE saves the result of SOUND-MIX into a new sound file. The new file is loaded in a SOUND box.
SOUND-SAVE saves the result of SOUND-MIX into a new sound file. The new file is loaded in a SOUND box.

Processing Functions

Below are listed the available sound processing functions. They all produce SOUND-DATA pointers.

  • SOUND-MIX : mixes two input sounds

  • SOUND-SEQ : concatenates two input sounds

  • SOUND-VOL : modifies the volume of an input sound

  • SOUND-CUT: extracts a specified interval out of an inpt sound

  • SOUND-FADE : adds a fade-in / fade-out effect to an input sound

  • SOUND-LOOP : repeats an input sound a number of times

  • SOUND-SILENCE : generates and empty (silent) sound of a given duration.

See the individual box documentation (press d) for more detailed descriptions.

Cascading Processes

The sound processing tools can also apply directly on SOUND-DATA pointers. It is therefore possible to apply sequences and combinations of effects and processes without using intermediary files.

SAVE-SOUND can be used eventually to store the final sound result in a sound file.

Cascading Sound Processing : An Algorithmic "Montage"

This example implements the following process :

  • 1 second of sound (between 2000 and 3000 ms) is extracted from SOUND1 ( sound-cut )
  • 500ms of silence ( sound-silence ) is appended after the result ( sound-seq )
  • SOUND2 is appended after the previous result ( sound-seq )
  • A global fade-in/fade out of 100ms is applied to the resulting sound ( sound-fade )
  • The final result is saved in the file "my-montage.aif" ( save-sound ) [ outfile autoamtically generates the output file pathname using the OM default out-file folder]

Automating Processing

The different programming facilities provided in OM (iterations, etc.) might be useful to automate sound processing on sound files and SOUND-DATA pointers.

REDUCE, MAPCAR...

In this example, we use higher-order functions in order to process lists of sound files and pointers.

  • mapcar sequencially cuts a segment from the sounds in the input list by calling SOUND-CUT. Note that the SOUND-CUT box is in lambda mode in order to be used as a functional argument to mapcar . The result is a list of SOUND-DATA pointers.
  • The Lisp function reduce allows to use a function (left input) to combine the successive pairs of elements in a list (right input). reduce applies SOUND-SEQ with elements 1 and two in the list, then to the result and element 3, and so forth. We use this function to automatically build a sequence from our list of SOUND-DATA. Note that SOUND-SEQ is also in in lambda mode ; This time it has two free inputs since reduce requires a binary operation to be supplied.
  • Finally the sequence is saved in a new sound file.

(More advanced processed could be imagined, and implemented with abstractions or using an omloop).

Voir aussi
page précédentepage suivante
A propos...(c) Ircam - Centre PompidouRéalisé avec Scenari