TEDLIUM 2

TEDLIUM-2 is a dataset used for evaluating automatic speech recognition (ASR) systems, containing transcribed audio recordings of TED Talks.

Papers using TEDLIUM 2 (17)

Boosting CTC-Based ASR Using LLM-Based Intermediate Loss Regularization2025 · 1 citesCJST: CTC Compressor based Joint Speech and Text Training for Decoder-Only ASR2024Distilling the Knowledge of BERT for CTC-based ASR2022 · 6 citesHierarchical Conditional End-to-End ASR with CTC and Multi-Granular Subword Units2021 · 3 citesMLP-ASR: Sequence-length agnostic all-MLP architectures for speech recognition2022 · 2 citesPrompting Large Language Models for Zero-Shot Domain Adaptation in Speech Recognition2023 · 2 citesImproving CTC-based ASR Models with Gated Interlayer Collaboration2022 · 1 citesASR Rescoring and Confidence Estimation with ELECTRA2021An Empirical Study of Language Model Integration for Transducer based Speech Recognition2022Non-autoregressive Error Correction for CTC-based ASR with Phone-conditioned Masked LM2022A Lexical-aware Non-autoregressive Transformer-based ASR Model2023Decoupled Structure for Improved Adaptability of End-to-End Models2023HypR: A comprehensive study for ASR hypothesis revising with a reference corpus2023Semi-Autoregressive Streaming ASR With Label Context2023Label-Synchronous Neural Transducer for Adaptable Online E2E Speech Recognition2023Weak Alignment Supervision from Hybrid Model Improves End-to-end ASR2023LV-CTC: Non-autoregressive ASR with CTC and latent variable models2024