Oral presentation in Underwater Technology 2019 Kaohsiung
Evaluating changes in the marine soundscape of an offshore wind farm via the machine learning-based source separation
Tzu-Hao Lin1, Hsin-Te Yang2, Jie- Mao Huang2, Chiou-Ju Yao3, Yung-Shun Lien4, Pei-Jung Wang4, Fang-Yu Hu4
1Japan Agency for Marine-Earth Science and Technology (JAMSTEC), Japan
2Observer Ecological Consultant, Taiwan
3Nature Museum of Natural Science, Taiwan
4Industrial Technology Research Institute, Taiwan
Investigating the ecological effects of offshore wind farms requires comprehensive surveys of marine ecosystem. Recently, the monitoring of marine soundscapes has been included in the rapid appraisals of geophysical events, marine fauna, and human activities. Machine learning is widely applied in acoustic research to improve the efficiency of audio processing. However, the use of machine learning to analyze marine soundscapes remains limited due to a general lack of human-annotated databases. In this study, we used unsupervised learning to recognize different sound sources underwater. We also quantified the temporal, spatial, and spectral variabilities of long-term underwater recordings collected near Phase I of the Formosa I wind farm. One source separation model was developed to recognize choruses made by fish and snapping shrimp, as well as shipping noise. Another model was developed to identify transient fish calls and echolocation clicks of marine mammals. Models were trained in an unsupervised manner using the periodicity-coded non-negative matrix factorization. After the sound sources were separated, events can be identified using Gaussian mixture models. Our information retrieval techniques facilitate future investigations of the spatiotemporal changes in marine soundscapes and allow to build an annotated database efficiently. The soundscape information can be used to evaluate the potential impacts of noise-generating activities on soniferous marine animals and their acoustic behavior before, during, and after the development of offshore wind farms.
Applying machine-learning based source separation techniques in the analysis of marine soundscapes
Tzu-Hao Lin1, Tomonari Akamatsu2, Yu Tsao3, Katsunori Fujikura1
1Department of Marine Biodiversity Research, Japan Agency for Marine-Earth Science and Technology
2National Research Institute of Fisheries Science, Japan Fisheries Research and Education Agency
3 Research Center for Information Technology Innovation, Academia Sinica
Long-term monitoring of underwater soundscapes provides us a large number of acoustic recordings to study a marine ecosystem. Characteristics of a marine ecosystem, such as the habitat quality, composition of marine fauna, and the level of human interference, may be analyzed using information relevant to environmental sound, biological sound, and anthropogenic sound. Supervised source separation techniques have been widely employed in speech and music separation tasks, but it may not be practical for the analysis of marine soundscapes due to the lack of a database that includes a large mount of paired pure and mixed signals. Even when the paired data is not available, different sound sources with unique spectral or temporal patterns may still be separated by apply semi-supervised or unsupervised learning algorithms. In this presentation, supervised and unsupervised source separation techniques will be demonstrated on long-term spectrograms of a marine soundscape. Separation performances under different levels of simultaneous source influence will also be discussed. In the future, more advanced techniques of source separation are necessary to facilitate the soundscape-based marine ecosystem sensing. An open database of marine soundscape will promote the development of machine learning-based source separation. Therefore, we will open acoustic data tested in this presentation on the Asian Soundscape to encourage the open science of marine soundscape.
Information retrieval from a soundscape by using blind source separation and clustering
Tzu-Hao Lin1, Yu Tsao2, Tomonari Akamatsu3, Mao-Ning Tuanmu4, Katsunori Fujikura1
1Department of Marine Biodiversity Research, Japan Agency for Marine-Earth Science and Technology
2Research Center for Information Technology Innovation, Academia Sinica
3National Research Institute of Fisheries Science, Japan Fisheries Research and Education Agency
4Biodiversity Research Center, Academia Sinica
Passive acoustic monitoring represents one of the remote sensing platforms of biodiversity. However, it remains challenging to retrieve meaningful biological information from a large amount of soundscape data when a comprehensive recognition database is not available. To overcome this issue, it is necessary to investigate the basic structure of a soundscape and subsequently retrieve biological information. The recent development of machine learning-based blind source separation techniques allow us to separate biological choruses and non-biological sounds appearing on a long-term spectrogram. After the blind source separation, the temporal-spatial changes of bioacoustic activities can be efficiently investigated by using a clustering algorithm. In this presentation, we will demonstrate the information retrieval in the forest and marine soundscapes. The separation result shows that in addition to biological information, we can also extract information relevant to weather patterns and human activities. Furthermore, the clustering result can be used to establish an audio library of nature soundscapes, which may facilitate the investigation of interactions among wildlife, climate change, and human development. In the future, the soundscape-based ecosystem monitoring will be feasible if we can integrate the soundscape information retrieval in a large-scale soundscape monitoring network.
Presentation in the session of Lidar and Passive observing sensors, Oceans’18 Kobe
Listening to the deep: Exploring marine soundscape variability by information retrieval techniques
Tzu-Hao Lin1,2, Yu Tsao2
1Department of Marine Biodiversity Research, Japan Agency of Marine-Earth Science and Technology, Yokosuka, Japan
2Research Center for Information Technology Innovation, Academia Sinica, Taipei, Taiwan
Information on the dynamics of the deep-sea ecosystem is essential for conservation management. The marine soundscape has been considered as an acoustical sensing
platform to investigate geophysical events, marine biodiversity, and human activities. However, analysis of the marine soundscape remains difficult because of the influence of simultaneous sound sources. In this study, we integrated machine learning-based information retrieval techniques to analyze the variability of the marine soundscape off northeastern Taiwan. A long-term spectral average was employed to visualize the longduration recordings of the Marine Cable Hosted Observatory (MACHO). Biotic and abiotic soundscape components were separated by applying periodicity-coded nonnegative matrix factorization. Finally, various acoustic events were identified
using k-means clustering. Our results show that the MACHO recordings of June 2012 contain multiple sound sources. Cetacean vocalizations, an unidentified biological chorus, environmental noise, and system noise can be accurately separated without an audio recognition database. Cetacean vocalizations were primarily detected at night, which is consistent with the detection results of two rule-based detectors. The unidentified biological chorus, ranging between 2 and 3 kHz, was primarily recorded between 7 p.m. and midnight during the studied period. On the basis of source separation, more acoustic events can be identified in the clustering result. The proposed
information retrieval techniques effectively reduce the difficulty in the analysis of marine soundscape. The unsupervised approach of source separation and clustering can improve the investigation regarding the temporal behavior and spectral characteristics of different sound sources. Based on the findings in the present study, we believe that variability of the deep-sea ecosystem can be efficiently investigated by combining the
soundscape information retrieval techniques and cabled hydrophone networks in the future.
Full text is available at: https://ieeexplore.ieee.org/document/8559307
2017/1/23-24 @ 高雄中山大學
The soundscape in shallow marine environment displays a high level of spatial variation due to the regional change of geophysical environment, biological community, and human activity. Many marine animals rely on sounds for orientation; therefore the soundscape has been hypothesized as one of the environmental indicators. Indo-Pacific humpback dolphins in western Taiwan waters are critically endangered. The sound perception is essential for humpback dolphins, which communicate through 3-15 kHz whistles and echolocate through ultrasonic clicks. However, the importance of soundscape for their habitat selection remains unclear. In this study, SM2+ recorders were deployed in Miaoli waters to collect long-duration underwater recordings. Echolocation clicks were automatically detected to identify the core habitat of humpback dolphins. The long-term spectral average reveals that the soundscape in Miaoli waters evidently changed among the diurnal cycle. The spectral characteristic varied between the core habitat and non-core habitat. The soundscape at the core habitat was characterized by the higher standard deviations and lower means of SPL in mid- and high-frequency range. It indicates that the nighttime chorus of croakers and the low-level of ambient sound in the daytime represent the classical soundscape at the core habitat. On the contrary, the croaker chorus was less prominent at the non-core habitat. Instead, snapping shrimp sounds dominated the local soundscape. The current results can help understand the soundscape change of humpback dolphin habitat during the future development and operation of offshore wind farms.
Keywords：Sousa chinensis, marine soundscape, biological sound, temporal and spatial variations, sound detector