Speech Enhancement on EARS-WHAM

Metric: SI-SDR (higher is better)

LeaderboardDataset

Loading chart...

Results

Submit a result

Sort:

#	Model↕	SI-SDR▼	Extra Data	Paper	Date↕	Code
1	Schrödinger Bridge	17.85	No	Schrödinger Bridge for Generative Speech Enhance...	2024-07-22	-
2	Conv-TasNet	16.93	No	Conv-TasNet: Surpassing Ideal Time-Frequency Mag...	2018-09-20	Code
3	Demucs v4	16.92	No	Hybrid Transformers for Music Source Separation	2022-11-15	Code
4	SGMSE+	16.78	No	Speech Enhancement and Dereverberation with Diff...	2022-08-11	Code
5	Schrödinger Bridge (PESQ loss)	16.29	No	Investigating Training Objectives for Generative...	2024-09-16	Code
6	CDiffuSE	8.35	No	Conditional Diffusion Probabilistic Model for Sp...	2022-02-10	Code

#1Schrödinger BridgeSOTA
17.85
SI-SDR· 2024-07-22
Schrödinger Bridge for Generative Speech Enhancement
#2Conv-TasNetSOTA
16.93
SI-SDR· 2018-09-20
Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separation Code
#3Demucs v4
16.92
SI-SDR· 2022-11-15
Hybrid Transformers for Music Source Separation Code
#4SGMSE+
16.78
SI-SDR· 2022-08-11
Speech Enhancement and Dereverberation with Diffusion-based Generative Models Code
#5Schrödinger Bridge (PESQ loss)
16.29
SI-SDR· 2024-09-16
Investigating Training Objectives for Generative Speech Enhancement Code
#6CDiffuSE
8.35
SI-SDR· 2022-02-10
Conditional Diffusion Probabilistic Model for Speech Enhancement Code