A Hands-On Comparison of DNNs for Dialog Separation Using Transfer Learning from Music Source Separation
| Autor: | Matteo Torcoli, Bernd Edler, Jouni Paulus, Martin Strauss |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
FOS: Computer and information sciences
Sound (cs.SD) Computer science Speech recognition SIGNAL (programming language) Separation (aeronautics) Degree (music) Computer Science - Sound Audio and Speech Processing (eess.AS) Application domain FOS: Electrical engineering electronic engineering information engineering Source separation Deep neural networks Dialog box Transfer of learning Electrical Engineering and Systems Science - Audio and Speech Processing |
| Zdroj: | Interspeech 2021. |
| DOI: | 10.21437/interspeech.2021-1418 |
| Popis: | This paper describes a hands-on comparison on using state-of-the-art music source separation deep neural networks (DNNs) before and after task-specific fine-tuning for separating speech content from non-speech content in broadcast audio (i.e., dialog separation). The music separation models are selected as they share the number of channels (2) and sampling rate (44.1 kHz or higher) with the considered broadcast content, and vocals separation in music is considered as a parallel for dialog separation in the target application domain. These similarities are assumed to enable transfer learning between the tasks. Three models pre-trained on music (Open-Unmix, Spleeter, and Conv-TasNet) are considered in the experiments, and fine-tuned with real broadcast data. The performance of the models is evaluated before and after fine-tuning with computational evaluation metrics (SI-SIRi, SI-SDRi, 2f-model), as well as with a listening test simulating an application where the non-speech signal is partially attenuated, e.g., for better speech intelligibility. The evaluations include two reference systems specifically developed for dialog separation. The results indicate that pre-trained music source separation models can be used for dialog separation to some degree, and that they benefit from the fine-tuning, reaching a performance close to task-specific solutions. accepted in INTERSPEECH 2021 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|