FADE is a simulation framework for auditory discrimination experiments. It is aimed to simulate, and hence predict the outcome of different experiments, from basic tone-in-noise detection experiments to the Matrix sentence test in fluctuating noise conditions, using an automatic speech recognition system with a universal set of parameters. The predictions with FADE are reference-free, which means no empirical data is required to perform predictions. The predictions match considerably well with empirical data . The approach is a generalized version of an approach to predict speech intelligibility . It also works with different languages , and was successfully used to predict the speech recognition performance of listeners with impaired hearing , and the benefit in speech recognition performance due to binaural noise reduction algorithms .
- Schädler, M. R., Warzybok, A., Ewert, S. D., and Kollmeier, B., "A simulation framework for auditory discrimination experiments: Revealing the importance of across-frequency processing in speech perception", Journal of the Acoustical Society of America, Volume 139, Issue 5, pp. 2708–2723 (2016). [link|download]
- Schädler, M. R., Warzybok, A., Hochmuth, S., and Kollmeier, B., "Matrix sentence intelligibility prediction using an automatic speech recognition system", International Journal of Audiology, Volume 54, Issue Supplement 2, pp 100–107 (2015). [link]
- Kollmeier, B., Schädler, M. R., Warzybok, A., Meyer, B., and Brand, T., "Sentence recognition prediction for hearing-impaired listeners in stationary and fluctuation noise with FADE: Empowering the Attenuation and Distortion concept by Plomp with a quantitative processing model", Trends in Hearing, Volume 20 (2016). [link]
- Schädler, M. R., Hülsmeier, D., Warzybok, A., Hochmuth, S., and Kollmeier, B. "Microscopic Multilingual Matrix Test Predictions Using an ASR-Based Speech Recognition Model" In Proc. INTERSPEECH, pp. 610–614 (2016).
- Schädler, M. R., Warzybok, A., & Kollmeier, B., "Objective Prediction of Hearing Aid Benefit Across Listener Groups Using Machine Learning: Speech Recognition Performance With Binaural Noise-Reduction Algorithms", Trends in Hearing, Volume 22 (2018). [link]
Licensing of the code: Dual-License
The simulation framework is licensed under both General Public License (GPL) version 3 and a proprietary license that can be arranged with us. In practical sense, this means:
- If you are developing Open Source Software (OSS) based on the FADE code, chances are you will be able to use it freely under GPL. But please double check here for OSS license compatibility with GPL.
- Alternatively, if you are unable to release your application as Open Source Software, you may arrange alternative licensing with us. Just send your inquiry to firstname.lastname@example.org to discuss this option.
FADE source code [repository]