Evaluating speech features with the minimal-pair ABX task: analysis of the classical MFC/PLP pipeline

Citations Over TimeTop 10% of 2013 papers

Abstract

We present a new framework for the evaluation of speech representations in zero-resource settings, that extends and complements previous work by Carlin, Jansen and Hermansky [1].In particular, we replace their Same/Different discrimination task by several Minimal-Pair ABX (MP-ABX) tasks.We explain the analytical advantages of this new framework and apply it to decompose the standard signal processing pipelines for computing PLP and MFC coefficients.This method enables us to confirm and quantify a variety of well-known and not-so-well-known results in a single framework.

Related Papers

• → HAIPipe: Combining Human-generated and Machine-generated Pipelines for Data Preparation(2023)19 cited
• → Design of a Novel Modular Serial Pipeline Inspection Robot(2023)4 cited
• → Development and Testing of a Field-Applied Coating for High-Temperature, Cathodically Protected Pipelines(2004)3 cited
• Three-Dimensional Numerical Simulation on Temperature Field of Crude Oil Pipeline and Products Pipeline Laid in One Ditch(2011)
• On Issues & Solutions of Separate System Drainage Pipeline Design(2013)