{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T09:55:27Z","timestamp":1773827727025,"version":"3.50.1"},"reference-count":28,"publisher":"IEEE","license":[{"start":{"date-parts":[[2024,4,14]],"date-time":"2024-04-14T00:00:00Z","timestamp":1713052800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,4,14]],"date-time":"2024-04-14T00:00:00Z","timestamp":1713052800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2024,4,14]]},"DOI":"10.1109\/icasspw62465.2024.10669911","type":"proceedings-article","created":{"date-parts":[[2024,9,10]],"date-time":"2024-09-10T18:23:22Z","timestamp":1725992602000},"page":"828-832","source":"Crossref","is-referenced-by-count":3,"title":["ISPA: Inter-Species Phonetic Alphabet for Transcribing Animal Sounds"],"prefix":"10.1109","author":[{"given":"Masato","family":"Hagiwara","sequence":"first","affiliation":[{"name":"Earth Species Project"}]},{"given":"Marius","family":"Miron","sequence":"additional","affiliation":[{"name":"Earth Species Project"}]},{"given":"Jen-Yu","family":"Liu","sequence":"additional","affiliation":[{"name":"Earth Species Project"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.7717\/peerj.13152"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1353\/nlh.1996.0031"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-acl.869"},{"key":"ref4","article-title":"A survey of large language models","author":"Zhao","year":"2023"},{"key":"ref5","article-title":"Fastspeech 2: Fast and high-quality endto-end text to speech","volume-title":"ICLR","author":"Ren"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU.2017.8269008"},{"key":"ref7","article-title":"Variable-rate discrete representation learning","author":"Dieleman","year":"2021"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2018.2869928"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1007\/0-387-32845-9"},{"key":"ref10","article-title":"PESTO: Pitch estimation with self-supervised transposition-equivariant objective","volume-title":"Proc. of ISMIR","author":"Riou"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2013.2271648"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10095642"},{"key":"ref13","article-title":"Neural discrete representation learning","volume-title":"NeurIPS","author":"van den Oord"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3133208"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054362"},{"key":"ref16","article-title":"Computer-coding the ipa: a proposed extension of sampa","author":"Wells","year":"1995"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10096686"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1145\/2733373.2806390"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1121\/2.0000358"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1038\/sdata.2017.143"},{"key":"ref21","article-title":"Cornell birdcall identification","year":"2020"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1016\/j.anbehav.2003.07.016"},{"key":"ref23","article-title":"HumBugDB: a large-scale acoustic mosquito dataset","volume-title":"Proc. of the NeurIPS: Track on Datasets and Benchmarks","author":"Kiskin"},{"key":"ref24","article-title":"Roberta: A robustly optimized BERT pretraining approach","author":"Liu","year":"2019"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.emnlp-demos.6"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3122291"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-698"},{"key":"ref28","article-title":"Scaling laws for neural language models","author":"Kaplan","year":"2020"}],"event":{"name":"2024 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)","location":"Seoul, Korea, Republic of","start":{"date-parts":[[2024,4,14]]},"end":{"date-parts":[[2024,4,19]]}},"container-title":["2024 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10625769\/10625780\/10669911.pdf?arnumber=10669911","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,9,11]],"date-time":"2024-09-11T17:29:41Z","timestamp":1726075781000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10669911\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,4,14]]},"references-count":28,"URL":"https:\/\/doi.org\/10.1109\/icasspw62465.2024.10669911","relation":{},"subject":[],"published":{"date-parts":[[2024,4,14]]}}}