GailBotPluginsTools

Overview

GailBotPluginsTools is a package designed to facilitate the creation and management of plugins for the GailBot platform. It simplifies the interaction with transcriptions, providing data handlers that manipulate XML markers and utterances.

Features

• Marker Utterance Dictionary: Efficiently manages utterances, their timings, and speaker information.
• Structured Interaction: Provides an outer layer that wraps around the marker utterance dictionary for enhanced functionality.
• Data Objects: Defines necessary data structures for handling utterances effectively.

Usage

The package includes several key modules:

marker_utterance_dict.py

• Class: MarkerUtteranceDict
• Purpose: Creates a dictionary to manage utterances, keeping track of speaker information, start and end times, and overlaps.
• Key Methods:
  • __init__: Initializes the dictionary with utterance data.
  • turn_criteria_overlaps: Checks if the difference between two utterances meets a defined threshold.

StructureInteract.py

• Class: StructureInteract
• Purpose: Interacts with the marker utterance dictionary and manages the output path for plugin results.
• Key Methods:
  • apply: The main driver function that processes the input data and populates the data structure.
  • parse_xml_to_gbpluginmethods: Converts XML data into a format usable by the GailBot Plugin methods.

data_objects.py

• Class: UttObj
• Purpose: Defines the structure for an utterance object, encapsulating its properties like start time, end time, speaker, and text.

Installation

To install the package, use the following command:

pip install gailbottools

Example

Here is a simple example of how to use the package:

from gailbot_plugin_tools import StructureInteract

# Sample XML data
xml_data = """
<transcript>
    <u speaker="Speaker1">
        <w start="0.0" end="0.5">Hello</w>
        <w start="0.6" end="1.0">World</w>
    </u>
</transcript>
"""

# Create an instance of StructureInteract
interactor = StructureInteract()
# Apply the XML data
interactor.apply(xml_data)

Contributing

Contributions are welcome! Please submit a pull request or open an issue for suggestions and enhancements.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Authors

• Hannah Shader
• Jason Wu
• Jacob Boyar
• Vivian Li
• Eva Caro
• Manuel Pena
• Phi Bui