Valentini

The Valentini dataset is used to evaluate the performance of speech enhancement models, specifically in the context of utilizing vector quantization techniques for improving speech denoising.

Papers using Valentini (4)

Multi-View Attention Transfer for Efficient Speech Enhancement2022 · 12 citesSpeech Enhancement Using Self-Supervised Pre-Trained Model and Vector Quantization2022 · 5 citesAccelerating RNN-based Speech Enhancement on a Multi-Core MCU with Mixed FP16-INT8 Post-Training Quantization2022Speech Enhancement with Multi-granularity Vector Quantization2023