Semi-supervised learning with deep neural networks for relative transfer function inverse regression

Ziteng Wang 1 Junfeng Li 1 Yonghong Yan 1 Emmanuel Vincent 2
2 MULTISPEECH - Speech Modeling for Facilitating Oral-Based Communication
Inria Nancy - Grand Est, LORIA - NLPKD - Department of Natural Language Processing & Knowledge Discovery
Abstract : Prior knowledge of the relative transfer function (RTF) is useful in many applications but remains little studied. In this paper, we propose a semi-supervised learning algorithm based on deep neural networks (DNNs) for RTF inverse regression, that is to generate the full-band RTF vector directly from the source-receiver pose (position and orientation). Two typical scenarios are discussed: training on labeled RTFs only, or on additional unlabeled RTFs. Both setups utilize the low-dimensional manifold property of RTF in stationary environments. With this property as an additional regulariza-tion term, a smooth mapping solution with respect to the manifold is obtained. Experimental simulations show that the proposed method achieves a lower mean prediction error than the free field model with few labeled RTFs, and the unlabeled RTFs are essential in improving the inverse regression performance.
Document type :
Conference papers
Complete list of metadatas

Cited literature [27 references]  Display  Hide  Download

https://hal.inria.fr/hal-01797886
Contributor : Emmanuel Vincent <>
Submitted on : Tuesday, May 22, 2018 - 8:31:53 PM
Last modification on : Wednesday, April 3, 2019 - 1:23:02 AM
Long-term archiving on : Tuesday, September 25, 2018 - 10:49:02 AM

File

wang_ICASSP18.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01797886, version 1

Citation

Ziteng Wang, Junfeng Li, Yonghong Yan, Emmanuel Vincent. Semi-supervised learning with deep neural networks for relative transfer function inverse regression. ICASSP 2018 – IEEE International Conference on Acoustics, Speech and Signal Processing, Apr 2018, Calgary, Canada. ⟨hal-01797886⟩

Share

Metrics

Record views

326

Files downloads

327