TY - GEN
T1 - Real-life voice activity detection with LSTM Recurrent Neural Networks and an application to Hollywood movies
AU - Eyben, Florian
AU - Weninger, Felix
AU - Squartini, Stefano
AU - Schuller, Bjorn
PY - 2013/10/18
Y1 - 2013/10/18
N2 - A novel, data-driven approach to voice activity detection is presented. The approach is based on Long Short-Term Memory Recurrent Neural Networks trained on standard RASTA-PLP frontend features. To approximate real-life scenarios, large amounts of noisy speech instances are mixed by using both read and spontaneous speech from the TIMIT and Buckeye corpora, and adding real long term recordings of diverse noise types. The approach is evaluated on unseen synthetically mixed test data as well as a real-life test set consisting of four full-length Hollywood movies. A frame-wise Equal Error Rate (EER) of 33.2% is obtained for the four movies and an EER of 9.6% is obtained for the synthetic test data at a peak SNR of 0 dB, clearly outperforming three state-of-the-art reference algorithms under the same conditions.
AB - A novel, data-driven approach to voice activity detection is presented. The approach is based on Long Short-Term Memory Recurrent Neural Networks trained on standard RASTA-PLP frontend features. To approximate real-life scenarios, large amounts of noisy speech instances are mixed by using both read and spontaneous speech from the TIMIT and Buckeye corpora, and adding real long term recordings of diverse noise types. The approach is evaluated on unseen synthetically mixed test data as well as a real-life test set consisting of four full-length Hollywood movies. A frame-wise Equal Error Rate (EER) of 33.2% is obtained for the four movies and an EER of 9.6% is obtained for the synthetic test data at a peak SNR of 0 dB, clearly outperforming three state-of-the-art reference algorithms under the same conditions.
KW - Long Short-Term Memory
KW - Neural Networks
KW - Speech Detection
KW - Voice Activity Detection
UR - http://www.scopus.com/inward/record.url?scp=84890443834&partnerID=8YFLogxK
U2 - 10.1109/ICASSP.2013.6637694
DO - 10.1109/ICASSP.2013.6637694
M3 - Conference contribution
AN - SCOPUS:84890443834
SN - 9781479903566
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 483
EP - 487
BT - 2013 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013 - Proceedings
T2 - 2013 38th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2013
Y2 - 26 May 2013 through 31 May 2013
ER -