Metraipolis › archive › dc_7299aeb0bf55

WaveGen: Truly End-to-End Waveform Generation via Internal Spectral Trajectory Learning — arXiv:2611.12527

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.12527
License (as archived): CC0-1.0
Archived: 2026-11-10T18:44:10.181Z
SHA-256 of the archived text: ec7d5504950cea7d029cdc52e186a13a3ce95cf7ac6707e4ad35b6289ff28297
Stable URL of this page: https://metraipolis.com/documento/dc_7299aeb0bf55

How to cite

WaveGen: Truly End-to-End Waveform Generation via Internal Spectral Trajectory Learning — arXiv:2611.12527 — arXiv (en), https://arxiv.org/abs/2611.12527 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7299aeb0bf55 (Metraipolis archive dc_7299aeb0bf55, sha256 ec7d5504950cea7d029cdc52e186a13a3ce95cf7ac6707e4ad35b6289ff28297)

BibTeX

@misc{dc7299aeb0bf55,
  title = {WaveGen: Truly End-to-End Waveform Generation via Internal Spectral Trajectory Learning — arXiv:2611.12527},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.12527},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7299aeb0bf55; sha256 ec7d5504950cea7d029cdc52e186a13a3ce95cf7ac6707e4ad35b6289ff28297},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.12527. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

This work proposes WaveGen, a fully end-to-end diffusion-based text-to-waveform framework based on Spectral Forcing, which jointly learns an internal spectral trajectory and the target waveform trajectory within a single model, enabling direct end-to-end text-to-waveform generation. Unlike recent TTS pipelines, WaveGen does not rely on separately trained or externally decoded intermediate representations, such as neural audio codecs, Mel-spectrogram vocoders, or VAE-based latent models. In addition, WaveGen performs internal text-speech alignment within a single model, eliminating the need for external alignment modules such as duration predictors. To preserve a truly end-to-end formulation, our framework further avoids dependence on self-supervised representations, such as BERT-based text embeddings and Wav2Vec 2.0-based semantic representations, as well as semantic distillation methods for training optimization. To achieve this, we carefully design Spectral Forcing, a model architecture with disentangled diffusion heads, and a training objective based on a multi-scale spectral $v$-loss. We further validate the scalability of end-to-end TTS models by scaling the model size from 0.1B to 0.6B parameters. Using a single architecture, open-source data, and one-stage training, without external modules such as duration predictor or REPA, WaveGen demonstrates fully end-to-end waveform generation achieving state-of-the-art performance on the Emilia training dataset. In particular, WaveGen achieves a WER of 1.47 and a SIM of 0.75 on the Seed-en benchmark. Source code and checkpoints are available at this https URL

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Sound (cs.SD) ; Artificial Intelligence (cs.AI); Audio and Speech Processing (eess.AS); Signal Processing (eess.SP).
Pagina del lavoro: https://arxiv.org/abs/2611.12527 — 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 2026-11-10T18:37:37.697Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).