WenetSpeech

Wenetspeech is a multilingual speech dataset used to evaluate the performance of speech-to-text models in aligning speech and text representations across different languages.

Papers using WenetSpeech (15)

PART: Progressive Alignment Representation Training for Multilingual Speech-To-Text with LLMs2025Exploring Cross-Utterance Speech Contexts for Conformer-Transducer Speech Recognition Systems2025LESS: Large Language Model Enhanced Semi-Supervised Learning for Speech Foundational Models Using in-the-wild Data2025Contextualized Automatic Speech Recognition with Dynamic Vocabulary Prediction and Activation2025Zipformer: A faster and better encoder for automatic speech recognition2023 · 28 citesNextformer: A ConvNeXt Augmented Conformer For End-To-End Speech Recognition2022 · 3 cites3M: Multi-loss, Multi-path and Multi-level Neural Networks for speech recognition2022 · 1 citesWenetSpeech4TTS: A 12,800-hour Mandarin TTS Corpus for Large Speech Generation Model Benchmark2024 · 1 citesAn Empirical Study of Language Model Integration for Transducer based Speech Recognition2022CB-Conformer: Contextual biasing Conformer for biased word recognition2023Research on an improved Conformer end-to-end Speech Recognition Model with R-Drop Structure2023CIF-T: A Novel CIF-based Transducer Architecture for Automatic Speech Recognition2023AutoPrep: An Automatic Preprocessing Framework for In-the-Wild Speech Data2023CUSIDE-T: Chunking, Simulating Future and Decoding for Transducer based Streaming ASR2024UCorrect: An Unsupervised Framework for Automatic Speech Recognition Error Correction2024