18062857. TESTING CASCADED DEEP LEARNING PIPELINES COMPRISING A SPEECH-TO-TEXT MODEL AND A TEXT INTENT CLASSIFIER simplified abstract (International Business Machines Corporation)

From WikiPatents
Jump to navigation Jump to search

TESTING CASCADED DEEP LEARNING PIPELINES COMPRISING A SPEECH-TO-TEXT MODEL AND A TEXT INTENT CLASSIFIER

Organization Name

International Business Machines Corporation

Inventor(s)

Swagatam Haldar of Kolkata (IN)

Diptikalyan Saha of Bangalore (IN)

Deepak Vijaykeerthy of Bangalore (IN)

Aniya Aggarwal of New Delhi (IN)

Nishtha Madaan of Gurgaon (IN)

TESTING CASCADED DEEP LEARNING PIPELINES COMPRISING A SPEECH-TO-TEXT MODEL AND A TEXT INTENT CLASSIFIER - A simplified explanation of the abstract

This abstract first appeared for US patent application 18062857 titled 'TESTING CASCADED DEEP LEARNING PIPELINES COMPRISING A SPEECH-TO-TEXT MODEL AND A TEXT INTENT CLASSIFIER

Simplified Explanation: The patent application describes a system and method for testing a cascaded pipeline using labeled speech data to evaluate the robustness of the pipeline.

Key Features and Innovation:

  • System includes memory, processor, input component, cascaded pipeline, and evaluation component.
  • Input component receives test case with label from labeled speech data.
  • Evaluation component feeds test case to cascaded pipeline and evaluates output for robustness.
  • Cascaded pipeline consists of first and second models, with the first model being different from the second model.

Potential Applications: This technology can be used in speech recognition systems, natural language processing applications, and machine learning models that require robustness testing.

Problems Solved: The technology addresses the need for efficient and accurate testing of cascaded pipelines in complex systems.

Benefits:

  • Improved accuracy and reliability of cascaded pipelines.
  • Streamlined testing process for evaluating robustness.
  • Enhanced performance of speech recognition and natural language processing systems.

Commercial Applications: Potential commercial applications include speech recognition software, virtual assistants, customer service chatbots, and automated transcription services.

Prior Art: Prior research in the field of machine learning, speech recognition, and natural language processing may provide insights into similar testing methods and systems.

Frequently Updated Research: Stay updated on advancements in machine learning algorithms, speech recognition technologies, and natural language processing techniques to enhance the efficiency and accuracy of testing cascaded pipelines.

Questions about Testing Cascaded Pipelines: 1. How does the evaluation component assess the robustness of the cascaded pipeline? 2. What are the potential challenges in implementing this technology in real-world applications?


Original Abstract Submitted

One or more systems, devices, computer program products and/or computer-implemented methods of use provided herein relate to a process to facilitate testing a cascaded pipeline. A system can comprise a memory that stores computer executable components, and a processor that executes the computer executable components stored in the memory, wherein the computer executable components can comprise an input component, a cascaded pipeline, and an evaluation component. The input component can receive a test case associated with a label from labeled speech data represented by waveform. The evaluation component can feed the test case to the cascaded pipeline to obtain an output of the cascaded pipeline. The evaluation component can evaluate a robustness of the cascaded pipeline by comparing the output of the cascaded pipeline and the label. The cascaded pipeline can include a first model and a second model, and the first model can be different than the second model.

Retrieved from "http://wikipatents.org/index.php?title=18062857._TESTING_CASCADED_DEEP_LEARNING_PIPELINES_COMPRISING_A_SPEECH-TO-TEXT_MODEL_AND_A_TEXT_INTENT_CLASSIFIER_simplified_abstract_(International_Business_Machines_Corporation)&oldid=73210"