Libri-2Mix
Emerging32papers using it
2022first seen
Libri-2Mix is a dataset used to evaluate Target Speaker Extraction (TSE) performance in mixed speech scenarios.
Papers using Libri-2Mix (32)
- SoloSpeech: Enhancing Intelligibility and Quality in Target Speech Extraction through a Cascaded Generative PipelineUnifying Diarization, Separation, and ASR with Multi-Speaker EncoderSEF-PNet: Speaker Encoder-Free Personalized Speech Enhancement with
Local and Global Contexts AggregationTowards Streaming Target Speaker Extraction via Chunk-wise Interleaved Splicing of Autoregressive Language ModelAlphaFlowTSE: One-Step Generative Target Speaker Extraction via Conditional AlphaFlowTripleC Learning and Lightweight Speech Enhancement for Multi-Condition Target Speech ExtractionMeanFlow-TSE: One-Step Generative Target Speaker Extraction with Mean FlowGenTSE: Enhancing Target Speaker Extraction via a Coarse-to-Fine Generative Language ModelA Study of the Scale Invariant Signal to Distortion Ratio in Speech Separation with Noisy ReferencesLightweight speech enhancement guided target speech extraction in noisy multi-speaker scenariosElevating Robust Multi-Talker ASR by Decoupling Speaker Separation and
Speech RecognitionSPMamba: State-space model is all you need in speech separationOn Data Sampling Strategies for Training Neural Network Speech
Separation ModelsSpeaker-Aware Mixture of Mixtures Training for Weakly Supervised Speaker
ExtractionAdapting self-supervised models to multi-talker speech recognition using
speaker embeddingsScaling strategies for on-device low-complexity source separation with
Conv-TasnetSQ-Whisper: Speaker-Querying based Whisper Model for Target-Speaker ASRUnifying Speech Enhancement and Separation with Gradient Modulation for
End-to-End Noise-Robust Speech SeparationWeakly-Supervised Speech Pre-training: A Case Study on Target Speech
RecognitionMC-SpEx: Towards Effective Speaker Extraction with Multi-Scale
Interfusion and Conditional Speaker ModulationMossFormer2: Combining Transformer and RNN-Free Recurrent Network for
Enhanced Time-Domain Monaural Speech SeparationInvestigating self-supervised learning for speech enhancement and
separationMVNet: Memory Assistance and Vocal Reinforcement Network for Speech
EnhancementAudioSlots: A slot-centric generative model for audio separationTarget Speech Extraction with Conditional Diffusion ModelSPGM: Prioritizing Local Features for enhanced speech separation
performanceProbing Self-supervised Learning Models with Target Speech ExtractionNoise-robust Speech Separation with Fast Generative CorrectionOn the effectiveness of enrollment speech augmentation for Target
Speaker ExtractionWanna hear your voice? A sample is all we need!Multi-Level Speaker Representation for Target Speaker ExtractionU-Mamba-Net: A highly efficient Mamba-based U-net style network for
noisy and reverberant speech separation