GOLF: A Singing Voice Synthesiser with Glottal Flow Wavetables and LPC Filters
This paper introduces GlOttal-flow LPC Filter (GOLF), a novel method for singing voice synthesis (SVS) that exploits the physical characteristics of the human voice using differentiable digital signal processing. GOLF employs a glottal model as the harmonic source and LPC filters to simulate the voc...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Ubiquity Press
2024-12-01
|
| Series: | Transactions of the International Society for Music Information Retrieval |
| Subjects: | |
| Online Access: | https://account.transactions.ismir.net/index.php/up-j-tismir/article/view/210 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841554873138544640 |
|---|---|
| author | Chin-Yun Yu György Fazekas |
| author_facet | Chin-Yun Yu György Fazekas |
| author_sort | Chin-Yun Yu |
| collection | DOAJ |
| description | This paper introduces GlOttal-flow LPC Filter (GOLF), a novel method for singing voice synthesis (SVS) that exploits the physical characteristics of the human voice using differentiable digital signal processing. GOLF employs a glottal model as the harmonic source and LPC filters to simulate the vocal tract, resulting in an interpretable and efficient synthesis approach. We show it is competitive with state-of-the-art singing voice vocoders, requiring fewer synthesis parameters and less memory to train, and runs an order of magnitude faster for inference. Additionally, we demonstrate that GOLF implicitly learns to model the phase components and formants of the human voice, having the potential to control and analyse singing voices in a differentiable manner. Our results highlight the effectiveness of incorporating the physical properties of the voice production mechanism into SVS and underscore the advantages of signal-processing-based approaches, which offer greater interpretability and efficiency in synthesis. |
| format | Article |
| id | doaj-art-827b47ee342147498830cf8a752e5350 |
| institution | Kabale University |
| issn | 2514-3298 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Ubiquity Press |
| record_format | Article |
| series | Transactions of the International Society for Music Information Retrieval |
| spelling | doaj-art-827b47ee342147498830cf8a752e53502025-01-08T08:41:56ZengUbiquity PressTransactions of the International Society for Music Information Retrieval2514-32982024-12-0171316–330316–33010.5334/tismir.210210GOLF: A Singing Voice Synthesiser with Glottal Flow Wavetables and LPC FiltersChin-Yun Yu0György Fazekas1Centre for Digital Music, Queen Mary University of LondonCentre for Digital Music, Queen Mary University of LondonThis paper introduces GlOttal-flow LPC Filter (GOLF), a novel method for singing voice synthesis (SVS) that exploits the physical characteristics of the human voice using differentiable digital signal processing. GOLF employs a glottal model as the harmonic source and LPC filters to simulate the vocal tract, resulting in an interpretable and efficient synthesis approach. We show it is competitive with state-of-the-art singing voice vocoders, requiring fewer synthesis parameters and less memory to train, and runs an order of magnitude faster for inference. Additionally, we demonstrate that GOLF implicitly learns to model the phase components and formants of the human voice, having the potential to control and analyse singing voices in a differentiable manner. Our results highlight the effectiveness of incorporating the physical properties of the voice production mechanism into SVS and underscore the advantages of signal-processing-based approaches, which offer greater interpretability and efficiency in synthesis.https://account.transactions.ismir.net/index.php/up-j-tismir/article/view/210singing voice synthesisdifferentiable digital signal processinglinear predictive codingglottal flowwavetable synthesis |
| spellingShingle | Chin-Yun Yu György Fazekas GOLF: A Singing Voice Synthesiser with Glottal Flow Wavetables and LPC Filters Transactions of the International Society for Music Information Retrieval singing voice synthesis differentiable digital signal processing linear predictive coding glottal flow wavetable synthesis |
| title | GOLF: A Singing Voice Synthesiser with Glottal Flow Wavetables and LPC Filters |
| title_full | GOLF: A Singing Voice Synthesiser with Glottal Flow Wavetables and LPC Filters |
| title_fullStr | GOLF: A Singing Voice Synthesiser with Glottal Flow Wavetables and LPC Filters |
| title_full_unstemmed | GOLF: A Singing Voice Synthesiser with Glottal Flow Wavetables and LPC Filters |
| title_short | GOLF: A Singing Voice Synthesiser with Glottal Flow Wavetables and LPC Filters |
| title_sort | golf a singing voice synthesiser with glottal flow wavetables and lpc filters |
| topic | singing voice synthesis differentiable digital signal processing linear predictive coding glottal flow wavetable synthesis |
| url | https://account.transactions.ismir.net/index.php/up-j-tismir/article/view/210 |
| work_keys_str_mv | AT chinyunyu golfasingingvoicesynthesiserwithglottalflowwavetablesandlpcfilters AT gyorgyfazekas golfasingingvoicesynthesiserwithglottalflowwavetablesandlpcfilters |