LibriSpeech-PC test-clean

The 'LibriSpeech-PC test-clean' is a subset of the LibriSpeech dataset used to evaluate automatic speech recognition (ASR) models, specifically focusing on their performance in producing formatted text with punctuation and capitalization.

Papers using LibriSpeech-PC test-clean (13)

CLEAR: Continuous Latent Autoregressive Modeling for High-quality and Low-latency Speech Synthesis2025 · 7 citesOptimising Neural Speech Codecs for 300bps Communication using Reinforcement Learning2026ClariCodec: Optimising Neural Speech Codes for 200bps Communication using Reinforcement Learning2026Enhancing Fully Formatted End-to-End Speech Recognition with Knowledge Distillation via Multi-Codebook Vector Quantization2025Quantizing Whisper-small: How design choices affect ASR performance2025HuBERT-VIC: Improving Noise-Robust Automatic Speech Recognition of Speech Foundation Model via Variance-Invariance-Covariance Regularization2025BR-ASR: Efficient and Scalable Bias Retrieval Framework for Contextual Biasing ASR in Speech LLM2025Pseudo-Autoregressive Neural Codec Language Models for Efficient Zero-Shot Text-to-Speech Synthesis2025Noisy Training Improves E2E ASR for the Edge2021 · 1 citesImproving Semi-supervised End-to-end Automatic Speech Recognition using CycleGAN and Inter-domain Losses2022Macro-block dropout for improved regularization in training end-to-end speech recognition models2022Dynamic Chunk Convolution for Unified Streaming and Non-Streaming Conformer ASR2023CTC-Assisted LLM-Based Contextual ASR2024