Getting started¶
Main window¶
The main window is divided into several areas:
The file manager (left panel) displays the hierarchical structure of the project. It contains folders for your corpus files, queries, data tables, analyses, scripts, research notes, and bookmarks.
The information panel (right panel) displays and lets you edit metadata about the file(s) currently selected in the file manager. When multiple files are selected, the information panel shows a batch editing view where you can add, edit, or remove properties for all selected files at once.
The tools panel (bottom) contains three tabs:
Console: an interactive command line for Phonometrica’s scripting engine.
Output: a read-only panel where results from acoustic measurements, script execution, and other operations are displayed.
IPA: a clickable IPA chart for inserting phonetic symbols into annotations, scripts, or any text field. Click on a symbol to insert it at the cursor in the currently focused text widget.
The viewer (center) is the main working area. It displays views as tabs, similar to web pages in a modern browser. The default start view provides quick-access buttons for common operations:
Open Project…: open an existing project file.
Add Files…: import sound files, annotations, or data files into the current project.
New Annotation…: create a new annotation from a sound file already in the project.
Analyze Data…: open a CSV file or concordance for statistical analysis.
Documentation: open the built-in help.
Below the quick-action buttons, a Recent projects list shows the projects you have opened recently (up to 10). Click on a project name to reopen it. The full path is shown as a tooltip.
All panels can be shown or hidden via the Window menu. You can also resize and rearrange them by dragging their borders.
Typical workflow¶
A typical workflow in Phonometrica involves the following steps:
Create a project and import your sound files and annotations.
Add metadata to your files using properties (e.g. Speaker, Gender, Dialect).
Annotate your speech data, or import existing annotations (e.g. Praat TextGrids).
Run queries to search for patterns in your annotations and/or extract acoustic measurements (formants, pitch, intensity).
Explore and filter the results in concordance or dataset views: recode categories, transform values, exclude outliers.
Fit statistical models in the analysis view: linear regression, mixed-effects models, GAMs.
Export your results to CSV or LaTeX for inclusion in publications.
Each of these steps is described in detail in the following sections of the documentation.
Corpus management¶
Several commands in the File menu let you import files into a project, either individually
(File > Add files to project...) or by importing a directory recursively
(File > Add content of directory to project...).
The logical structure of a project is independent from the
physical organization of the files on your computer: once files
have been added to a project, they can be moved around, merged into new
directories or removed without affecting the files on disk. Phonometrica supports
several annotation formats, including TextGrid (Praat) and its own native XML-based
format. It also supports a number of audio formats, including WAV,
AIFF and FLAC (the exact number of supported formats depends on the
platform). By default, Phonometrica will try to automatically bind an
annotation and a sound file if they have the same base name but a
different extension. If the names differ, it is possible to bind them
manually, by right-clicking on them and choosing the corresponding
option in the context menu, or semi-automatically using the
Import metadata... feature from the file menu or using the scripting
engine.
Properties¶
The hierarchical organization of a project is a matter of pure convenience to the user and is irrelevant for Phonometrica. Instead, the program relies on metadata to keep files organized internally and to perform queries. File names represent the most basic type of metadata and for small projects (containing a dozen of files or so) this may be all that is needed. When you need to sort and organize a larger collection of files, Phonometrica offers a flexible mechanism called properties. A property is a typed key/value pair. Each file can be tagged with an arbitrary number of properties: the key represents a category, which is always a text string, and the value may be either Boolean, textual or numeric. Typical examples of properties would be Speaker (where each unique speaker identifier represents a distinct value, for example 11ajp1) and Gender (with the values Male and Female). Properties can be managed via the information panel when a file is selected.
When multiple files are selected in the file manager, the information panel displays a batch editing view. This shows a union table of all properties across the selected files, with columns for the property category, type (Boolean, text, or numeric), value, and a coverage count indicating how many of the selected files have each property. If a property has different values across the selected files, the value column shows (mixed). You can add, edit, or remove properties for all selected files at once. If the selected files have different descriptions, a checkbox lets you choose whether to overwrite them all with a new common description.
In addition to properties, each file can be annotated with a description, a free-form string which can be used to store any kind of information, and which is also exposed to the search engine to filter files.
Import metadata¶
Phonometrica lets you import metadata from a CSV file, which is a tabular format where each line
represents a row and values in a row are separated by a separator character. When you click on
Import metadata... in the File > Import menu, a dialog opens and asks you to choose
a file as well as the separator character. The semicolon is used by default, since this is the separator used by
Excel when it exports a spreadsheet to CSV. The file must have the following structure:
the first line must be a header: the first field is ignored, and the following fields correspond to property categories.
the first column must contain the name of the file to which you want to add metadata; the remaining columns contain values for the corresponding categories. Cells can be left empty for files that don’t have a given property.
The cells in the first column represent file names as they are normally displayed by your operating system. For instance,
if you want to tag the file C:\Test\loc1f26.wav, i.e. a file named loc1f26.wav located in a directory named
Test on the C: drive (on Windows), you should only provide loc1f26.wav as the file name. (Bear in mind that matching
is case sensitive, so LOC1F26.WAV would not work in this case.) In addition, Phonometrica allows you to use regular expressions
to define a file name pattern instead of an exact file name. To use a regular expression, the file name pattern must start with ^
(the metacharacter that indicates the beginning of a string) and end with $ (the metacharacter that indicates the end of a string). You
can use any valid regular expression supported by Phonometrica, including capturing parentheses. You can refer to the whole matched pattern or to any subgroup in subsequent fields, using either %% or the placeholders %1 to %9, respectively.
By default, properties are assigned the type text. You can add the suffixes .bool, .num or .text to the category in
the header line to indicate that the property is a Boolean, a number or a text string, respectively. In addition, there are two special fields for columns: %SOUND% and %DESCRIPTION%. %SOUND% allows you
to provide the full path of a sound file for an annotation. If the sound file exists and the file is indeed an
annotation, Phonometrica will bind the annotation to the sound; otherwise, the value will be dismissed. %DESCRIPTION%
allows you to set the description field of a file.