Improved Speech Enhancement with the Wave-U-Net
By Craig Macartney, Tillman Weyde. 2018
194 citations recorded by the source corpus.
We study the use of the Wave-U-Net architecture for speech enhancement, a model introduced by Stoller et al for the separation of music vocals and accompaniment. This end-to-end learning method for audio source separation operates directly in the time domain, permitting the integrated modelling of phase information and being able to take large temporal contexts into account. Our experiments show that the proposed method improves several metrics, namely PESQ, CSIG, CBAK, COVL and SSNR, over the state-of-the-art with respect to the speech enhancement task on the Voice Bank corpus (VCTK) dataset. We find that a reduced number of hidden layers is sufficient for speech enhancement in comparison to the original system designed for singing voice separation in music. We see this initial result as an encouraging signal to further explore speech enhancement in the time-domain, both as an end in itself and as a pre-processing step to speech recognition systems.
Subfields: Artificial Intelligence, Signal Processing
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Yossi Adi
- Gabriel Synnaeve
- Alexandre D'efossez
- Ritwik Giri
- Umut Isik
- A. Krishnaswamy
- Jean-Marc Valin
- Wei-Ning Hsu
In Signal Processing
- Yossi Adi
- Gabriel Synnaeve
- Alexandre D'efossez
- Ritwik Giri
- Umut Isik
- A. Krishnaswamy
- Karim Helwani
- Neerad Phansalkar
Citations in the corpus
- Real Time Speech Enhancement in the Waveform Domain (2020, background): “Over the last years, there has been a growing interest towards DNN based methods for speech enhancement [34, 35, 36, 37, 20, 38, 39, 22, 40, 7, 41, 8, 42, 21, 37].”
- PoCoNet: Better Speech Enhancement with Frequency-Positional Embeddings, Semi-Supervised Conversational Data, and Biased Loss (2020, methodology): “Recent phase-aware models use a complex ratio mask instead of magnitude gain [6, 7], while other approaches work directly in the waveform domain [8, 9, 10].”
- Separate Anything You Describe (2023, unknown): “To make a fair comparison with previous speech enhancement systems [6], [7], [69], [70], we resample all audio clips to 16 kHz.”