Version 1.5.0

• Introduced a rudimental Graphical User Interface (GUI)

• Customizable settings including preferred encodings, CSV separators, and GUI layout options

Corpus Loading

• Integration of a speaker JSON file for defining speaker-independent variables

• Capability to load pre-existing datasets

• Visual representation of the loaded dataset and defined variables

Dataset Creation

• Option to continue from where you left off in previous sessions

• Structured error reporting that allows continuous execution. Errors are categorized by severity: info, warning, and error.

• In-built stop button for ceasing dataset generation when needed

• GUI window to define regex rules, n-gram context, and data generation fields

• Optimized dataset creation process, now 25 times faster

Automatic Analysis

• Integrated window for automatic analysis tasks

• Support for various analyses including:

  • k-means clustering

  • Pairwise frequency computation

  • Cross-tabulation

  • t-test statistical analysis

Version 1.0.0 

Corpus Loading

• Support for loading multiple corpus files

• Compatibility with International Phonetic Alphabet (IPA) and UTF-16 encoding

• Recognition of different speakers within a single file

Variable Definition

• Define custom variables depending on specific research questions

• Support for both independent (e.g. age, gender, nationality) and dependent variables as JSON files

• Flexibility with broad annotation rules using regular expressions

Dataset Creation

• Creation of structured datasets in CSV format

• Automated data cleaning and preprocessing for extracted annotations

• Detailed output files, including annotation information, missed annotations, and corpus statistics

• Efficient error-tracking for consistency and accuracy