Description
Iterative Pseudo-Labeling (IPL) is a semi-supervised algorithm for speech recognition which efficiently performs multiple iterations of pseudo-labeling on unlabeled data as the acoustic model evolves. In particular, IPL fine tunes an existing model at each iteration using both labeled data and a subset of unlabeled data.
Papers Using This Method
IPL: Leveraging Multimodal Large Language Models for Intelligent Product Listing2024-10-22FanCric : Multi-Agentic Framework for Crafting Fantasy 11 Cricket Teams2024-10-02Speaker-IPL: Unsupervised Learning of Speaker Characteristics with i-Vector based Pseudo-Labels2024-09-16On the Role of Reflectarrays for Interplanetary Links2024-05-01Effective and anatomical connectivity of the dorso-central insula during the processing of action forms2023-10-22Nonparametric mixed logit model with market-level parameters estimated from market share data2023-09-22Inverse Preference Learning: Preference-based RL without a Reward Function2023-09-21ORMIR_XCT: A Python package for high resolution peripheral quantitative computed tomography image processing2023-09-08Popularity Debiasing from Exposure to Interaction in Collaborative Filtering2023-05-09Zero-shot Generative Model Adaptation via Image-specific Prompt Learning2023-04-06A Robust Approach for the Decomposition of High-Energy-Consuming Industrial Loads with Deep Learning2022-03-11PL-Net: Progressive Learning Network for Medical Image Segmentation2021-10-27Prediction of IPL Match Outcome Using Machine Learning Techniques2021-09-30Efficient Feature Representations for Cricket Data Analysis using Deep Learning based Multi-Modal Fusion Model2021-08-16Interactive Prototype Learning for Egocentric Action Recognition2021-01-01SlimIPL: Language-Model-Free Iterative Pseudo-Labeling2020-10-22Iterative Pseudo-Labeling for Speech Recognition2020-05-19