Are Recent Deep Learning-Based Speech Enhancement Methods
Ready to Confront Real-World Noisy Environments?
Paper link:
TBD
Github page:
https://github.com/candyolivia/is2024_deep_enhancement
Candy Olivia Mawalim, Shogo Okada, and Masashi Unoki
Japan Advanced Institute of Science and Technology
Samples:
CEC2 Development Set (CH1)
Unprocessed Data:
Development Set Mix
S06022_mix_CH1
S06088_mix_CH1
S06221_mix_CH1
DeepFilterNet3 (Fine-tune)
S06022_mix_CH1_DeepFilterNet3
S06088_mix_CH1_DeepFilterNet3
S06221_mix_CH1_DeepFilterNet3
Denoiser (H=48) (Fine-tune)
S06022_mix_CH1_denoiser48
S06088_mix_CH1_denoiser48
S06221_mix_CH1_denoiser48