Metraipolis › archive › dc_340ac4cd18ce
Unsupervised Instantaneous Phase and Frequency Tracking by Inverse Voice Synthesis — arXiv:2610.03058
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.03058
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T11:52:25.672Z
- SHA-256 of the archived text
7d797aa16cc54cf085e5cea1b67b0c057b16bc89864db13a4bd652d967deda39- Stable URL of this page
https://metraipolis.com/documento/dc_340ac4cd18ce
How to cite
Unsupervised Instantaneous Phase and Frequency Tracking by Inverse Voice Synthesis — arXiv:2610.03058 — arXiv (en), https://arxiv.org/abs/2610.03058 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_340ac4cd18ce (Metraipolis archive dc_340ac4cd18ce, sha256 7d797aa16cc54cf085e5cea1b67b0c057b16bc89864db13a4bd652d967deda39)
BibTeX
@misc{dc340ac4cd18ce,
title = {Unsupervised Instantaneous Phase and Frequency Tracking by Inverse Voice Synthesis — arXiv:2610.03058},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.03058},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_340ac4cd18ce; sha256 7d797aa16cc54cf085e5cea1b67b0c057b16bc89864db13a4bd652d967deda39},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.03058. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Knowledge-driven neural vocoders struggle to learn reliable fundamental frequency end-to-end, because spectral objectives provide weak supervision of periodic structure and lack phase information. We address this with a source-filter model whose alias-free additive source makes the instantaneous phase of the glottal cycle explicit; differentiating it yields the instantaneous frequency, and thus $F_0$, without an external tracker. Waveform error supervises only the deterministic harmonic path, while a spectral loss covers the full signal. On M4Singer and LM-SSD, the reconstruction is phase-aligned, reaching a signal-to-reconstruction-error ratio of 8.1 dB, while neural baselines remain negative. However, GOLF, given an external $F_0$, still reaches lower spectral distortion. On LM-SSD, the recovered $F_0$ attains the highest overall accuracy of any method tested, including supervised neural pitch trackers applied off the shelf, and the glottal closure instants come within 0.53 points of REAPER's identification rate, without any $F_0$ label. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-05 in eess.AS). Categorie arXiv: Audio and Speech Processing (eess.AS) ; Sound (cs.SD). Pagina del lavoro: https://arxiv.org/abs/2610.03058 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-06T10:15:42.620Z dalla pagina https://arxiv.org/catchup/eess.AS/2026-10-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).