Playing a part: speaker verification at the movies
The goal of this work is to investigate the performance of popular speaker recognition models on speech segments from movies, where often actors intentionally disguise their voice to play a character. We make the following three contributions: (i) We collect a novel, challenging speaker recognition...
| Main Authors: | , , , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
IEEE
2021
|
| _version_ | 1797057469795532800 |
|---|---|
| author | Brown, A Huh, J Nagrani, A Chung, JS |
| author_facet | Brown, A Huh, J Nagrani, A Chung, JS |
| author_sort | Brown, A |
| collection | OXFORD |
| description | The goal of this work is to investigate the performance of popular speaker recognition models on speech segments from movies, where often actors intentionally disguise their voice to play a character. We make the following three contributions: (i) We collect a novel, challenging speaker recognition dataset called VoxMovies, with speech for 856 identities from almost 4000 movie clips. VoxMovies contains utterances with varying emotion, accents and background noise, and therefore comprises an entirely different domain to the interview-style, emotionally calm utterances in current speaker recognition datasets such as VoxCeleb; (ii) We provide a number of domain adaptation evaluation sets, and benchmark the performance of state-of-the-art speaker recognition models on these evaluation pairs. We demonstrate that both speaker verification and identification performance drops steeply on this new data, showing the challenge in transferring models across domains; and finally (iii) We show that simple domain adaptation paradigms improve performance, but there is still large room for improvement. |
| first_indexed | 2024-03-06T19:36:49Z |
| format | Conference item |
| id | oxford-uuid:1f57f566-84db-449c-ab7b-84fce34ae611 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-06T19:36:49Z |
| publishDate | 2021 |
| publisher | IEEE |
| record_format | dspace |
| spelling | oxford-uuid:1f57f566-84db-449c-ab7b-84fce34ae6112022-03-26T11:21:14ZPlaying a part: speaker verification at the moviesConference itemhttp://purl.org/coar/resource_type/c_5794uuid:1f57f566-84db-449c-ab7b-84fce34ae611EnglishSymplectic ElementsIEEE2021Brown, AHuh, JNagrani, AChung, JSThe goal of this work is to investigate the performance of popular speaker recognition models on speech segments from movies, where often actors intentionally disguise their voice to play a character. We make the following three contributions: (i) We collect a novel, challenging speaker recognition dataset called VoxMovies, with speech for 856 identities from almost 4000 movie clips. VoxMovies contains utterances with varying emotion, accents and background noise, and therefore comprises an entirely different domain to the interview-style, emotionally calm utterances in current speaker recognition datasets such as VoxCeleb; (ii) We provide a number of domain adaptation evaluation sets, and benchmark the performance of state-of-the-art speaker recognition models on these evaluation pairs. We demonstrate that both speaker verification and identification performance drops steeply on this new data, showing the challenge in transferring models across domains; and finally (iii) We show that simple domain adaptation paradigms improve performance, but there is still large room for improvement. |
| spellingShingle | Brown, A Huh, J Nagrani, A Chung, JS Playing a part: speaker verification at the movies |
| title | Playing a part: speaker verification at the movies |
| title_full | Playing a part: speaker verification at the movies |
| title_fullStr | Playing a part: speaker verification at the movies |
| title_full_unstemmed | Playing a part: speaker verification at the movies |
| title_short | Playing a part: speaker verification at the movies |
| title_sort | playing a part speaker verification at the movies |
| work_keys_str_mv | AT browna playingapartspeakerverificationatthemovies AT huhj playingapartspeakerverificationatthemovies AT nagrania playingapartspeakerverificationatthemovies AT chungjs playingapartspeakerverificationatthemovies |