13
0
livetrax/manual/xml/using_existing_audio.xml

311 lines
11 KiB
XML
Raw Normal View History

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE chapter PUBLIC "-//OASIS//DTD DocBook XML V4.4//EN"
"http://www.oasis-open.org/docbook/xml/4.4/docbookx.dtd">
<chapter id="ch-using-existing-audio">
<title>Using Existing Audio</title>
<para>There are two primary ways to bring data into Ardour: recording it
within a session from a live sound source or importing pre-existing audio
files. This section covers the various ways to import audio into a
session.</para>
<section id="importing-and-embedding">
<title>Importing and Embedding</title>
<para>Importing and embedding are two different methods of using existing
audio files on your computer (or network file system) within a session.
They differ in one key respect:</para>
<variablelist>
<varlistentry>
<term>Importing</term>
<listitem>
<para>An existing audio file is copied to the session's sounds
folder, and is converted into the session's native format (WAVE or
Broadcast WAVE depending on your choice) and sample rate. At this
time, no control over the conversion process is offered. If sample
rate conversion is required, it will be carried out at the highest
quality that Ardour can provide. This means that it can be rather
slow (many minutes to import an audio file lasting a few
minutes).</para>
</listitem>
</varlistentry>
<varlistentry>
<term>Embedding</term>
<listitem>
<para>An existing audio file is used as a the source for a region,
but is not copied or modified in any way.</para>
</listitem>
</varlistentry>
</variablelist>
</section>
<section id="supported-external-audio-file-formats">
<title>Supported External Audio File Formats</title>
<para>The list of audio file formats that Ardour can import/embed is quite
long. It is based on the functionality offered by libsndfile, an excellent
and widely used software library by Australian programmer Erik de Castro
Lopo. As libsndfile's capabilities expand, so will Ardour's abilities to
import (and export) new formats. Ogg/Vorbis (an excellent, unpatented and
license free audio compression format similar to MP3) is planned for the
near future. Currently, supported formats include:</para>
<itemizedlist>
<listitem>
<para>Microsoft WAV</para>
</listitem>
<listitem>
<para>SGI/Apple AIFF/AIFC</para>
</listitem>
<listitem>
<para>Sun AU/Snd</para>
</listitem>
<listitem>
<para>Raw (headerless)</para>
</listitem>
<listitem>
<para>Paris Audio File (PAF)</para>
</listitem>
<listitem>
<para>Commodore IFF/SVX</para>
</listitem>
<listitem>
<para>Sphere/NIST WAV</para>
</listitem>
<listitem>
<para>IRCAM SF</para>
</listitem>
<listitem>
<para>Creative VOC</para>
</listitem>
<listitem>
<para>SoundForge W64</para>
</listitem>
<listitem>
<para>GNU Octave MAT4.4</para>
</listitem>
<listitem>
<para>Portable Voice Format</para>
</listitem>
<listitem>
<para>Fasttracker 2 XI</para>
</listitem>
<listitem>
<para>HMM Tool Kit HTK</para>
</listitem>
</itemizedlist>
<para>Sample encodings supported include:</para>
<itemizedlist>
<listitem>
<para>Unsigned and signed 8, 16, 24 and 32 bit PCM</para>
</listitem>
<listitem>
<para>IEEE 32 and 64 floating point</para>
</listitem>
<listitem>
<para>U-LAW</para>
</listitem>
<listitem>
<para>A-LAW</para>
</listitem>
<listitem>
<para>IMA ADPCM</para>
</listitem>
<listitem>
<para>MS ADPCM</para>
</listitem>
<listitem>
<para>GSM 6.10</para>
</listitem>
<listitem>
<para>G721/723 ADPCM</para>
</listitem>
<listitem>
<para>12/16/24 bit DWVW</para>
</listitem>
<listitem>
<para>OK Dialogic ADPCM</para>
</listitem>
<listitem>
<para>8/16 DPCM</para>
</listitem>
</itemizedlist>
</section>
<section id="using-audio-files">
<title>Using audio files as tracks or regions?</title>
<para>When you want to use existing audio files in an Ardour session, the
first choice you need to make is whether you want to bring the files in as
tracks or as new regions. Consider the two following scenarios:</para>
<itemizedlist>
<listitem>
<para>you have an 8 track recording of existing material, with 1 audio
file per track</para>
</listitem>
<listitem>
<para>you have a sample library containing 500 small audio
files</para>
</listitem>
</itemizedlist>
<para>In the first case, your goal is probably to have 8 tracks (at
least), with each track containing a single audio file. In the second
case, its a lot more likely that you simply want to be able to use any of
the samples easily, but do not want any tracks created as a direct result
of the import/embed. It is very important that you understand this
distinction: many new users think there should be a "simple" way to import
existing audio without understanding that the goal of importing/embedding
is not always the same.</para>
<para>Ardour provides two different options when importing. You can
import/embed audio files as new tracks, or you can import/embed them into
the region list, where they will be available as regions to put into new
or existing tracks. You can also insert import/embed audio files directly
into an existing track.</para>
</section>
<section id="how-to-import-embed">
<title>How to import/embed</title>
<para>There are three pathways for importing/embedding an audio file into
a session. One is initiated from the File menu, one is initiated from the
track context menu, and one is initiated from the region list. These
methods are all equivalent: they open the file import dialog. Once the
dialog is open, you can choose to add new audio as new tracks, as regions
in the region list, or as audio in the selected track. You can change this
behavior once the dialog is opened.</para>
<para><caution>
<para>If you want to import a file directly into an existing track,
you must have the track selected in the editor before beginning the
import. </para>
</caution>If you use Nautilus as your file manager, you can easily
import files into your project by dragging them onto the desired track,
then releasing the mouse button. The file will then be <link
linkend="gt-embed">embedded</link> into your session.</para>
</section>
<section id="how-to-embed-a-file">
<title>Working with Tags</title>
<para>A "tag" is bit of information, or <emphasis>metadata</emphasis>,
that is associated with a data file. Specifically, tags are keywords or
terms that you feel have some relevance to a particular soundfile. Ardour
can store these tags in a seachable database so that you can quickly
search for sounds based on the tags that you have assigned to them. For
example you can assign the term "120bpm" to a sound, and then when you
search for this tag, the file will appear in the search list. Tags are
independent of the filename or anything else about the file. Tags, and the
file paths that they are associated with, are stored in a file called
"sfdb" in your Ardour user folder. </para>
<section id="embedding-via-the-region-list">
<title>Adding tags to a file</title>
<orderedlist>
<listitem>
<para>Select a file in the import dialog.</para>
</listitem>
<listitem>
<para>Enter the tags in the box to the right labeled "Tags".</para>
</listitem>
</orderedlist>
<para>Tags are stored when the input box loses focus, there is no reason
to explicitly "save" them.</para>
</section>
<section id="embedding-from-a-track">
<title>Searching for files using tags</title>
<orderedlist>
<listitem>
<para>Select the "Search Tags" tab on the import dialog.</para>
</listitem>
<listitem>
<para>Enter the tag(s) to search for and press "Search".</para>
</listitem>
</orderedlist>
<para>Files which have been "tagged" with the input terms will appear in
the results window. You can audition these files and apply tags to them
from this window.</para>
</section>
</section>
<section id="how-to-embed-a-file">
<title>Searching Freesound for soundfiles (optional)</title>
<para>Freesound (http://freesound.iua.upf.edu/) is an online database with
thousands of soundfiles which can be freely used in many projects (visit
Freesound for the exact licensing terms). Ardour has an option to download
files directly from freesound directly from the import dialog. Freesound
files are tagged with metadata which you can use to search for relevant
sounds.</para>
<para>To enable the Freesound search engine inside Ardour, you must build
Ardour with the option FREESOUND=yes, and you must have the "curl" library
installed. This is currently optional but may later become a standard
feature of Ardour.</para>
<para>To download files from Freesound, you must first register for a
username and password on the freesound website. Enter this information
into the text entry boxes, enter your tags to search for, and then click
"Start Downloading". Ardour will begin downloading files, and the "Start
Downloading" button will change to say "Cancel". Clicking the "Cancel"
button will wait until the currently downloading file is finished, then
the button will switch back to "Start Downloading". The button will also
say "Start Downloading" once Ardour has downloaded all of the files that
match the given tag. In the future there may be better indication of the
file progress and better filtering on the filetypes that are
downloaded.</para>
<para>If you later search for the same terms, you will see that the files
appear in the list more quickly, because any files in the search set that
are already downloaded won't have to be downloaded again.</para>
<para>Sounds that are downloaded from Freesound will automatically be
given tags in Ardour that match their tags in the Freesound database. This
means that once they are downloaded, Ardour can search for the local files
very quickly using the "Search Tags" tab.</para>
</section>
<!--
<xi:include xmlns:xi="http://www.w3.org/2001/XInclude"
href="Some_Subsection.xml" />
-->
</chapter>