4 papers
Training DeepFilterNet with Accurate Room Acoustic Simulations Improves Single-Channel Speech Enhancement
Alessia Milo, Georg Götz, Steinar Guðjónsson +3
We investigate how the realism of synthetic room impulse response (RIR) datasets affects the training of DeepFilterNet3 for single-channel speech enhancement. We compare a DNS4 ima…
Improving multichannel speech enhancement through accurate room-acoustic simulations
Georg Götz, Alessia Milo, Steinar Guðjónsson +3
Room-acoustic simulations are widely used to augment training data for deep-learning-based speech enhancement. While most pipelines rely on simplified geometrical acoustics, wave-b…
Treble10: A high-quality dataset for far-field speech recognition, dereverberation, and enhancement
Sarabeth S. Mullins, Georg Götz, Eric Bezzam +2
Accurate far-field speech datasets are critical for tasks such as automatic speech recognition (ASR), dereverberation, speech enhancement, and source separation. However, current d…
Room-acoustic simulations as an alternative to measurements for audio-algorithm evaluation
Georg Götz, Daniel Gert Nielsen, Steinar Guðjónsson +1
Audio-signal-processing and audio-machine-learning (ASP/AML) algorithms are ubiquitous in modern technology like smart devices, wearables, and entertainment systems. Development of…