Lightweight End-to-end Text-to-speech Synthesis for low resource on-device applications
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915633993613312 |
|---|---|
| author | Vecino, Biel Tura Gabryś, Adam Mątwicki, Daniel Pomirski, Andrzej Iddon, Tom Cotescu, Marius Lorenzo-Trueba, Jaime |
| author_facet | Vecino, Biel Tura Gabryś, Adam Mątwicki, Daniel Pomirski, Andrzej Iddon, Tom Cotescu, Marius Lorenzo-Trueba, Jaime |
| contents | Recent works have shown that modelling raw waveform directly from text in an end-to-end (E2E) fashion produces more natural-sounding speech than traditional neural text-to-speech (TTS) systems based on a cascade or two-stage approach. However, current E2E state-of-the-art models are computationally complex and memory-consuming, making them unsuitable for real-time offline on-device applications in low-resource scenarios. To address this issue, we propose a Lightweight E2E-TTS (LE2E) model that generates high-quality speech requiring minimal computational resources. We evaluate the proposed model on the LJSpeech dataset and show that it achieves state-of-the-art performance while being up to $90\%$ smaller in terms of model parameters and $10\times$ faster in real-time-factor. Furthermore, we demonstrate that the proposed E2E training paradigm achieves better quality compared to an equivalent architecture trained in a two-stage approach. Our results suggest that LE2E is a promising approach for developing real-time, high quality, low-resource TTS applications for on-device applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_07701 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Lightweight End-to-end Text-to-speech Synthesis for low resource on-device applications Vecino, Biel Tura Gabryś, Adam Mątwicki, Daniel Pomirski, Andrzej Iddon, Tom Cotescu, Marius Lorenzo-Trueba, Jaime Sound Artificial Intelligence Audio and Speech Processing Recent works have shown that modelling raw waveform directly from text in an end-to-end (E2E) fashion produces more natural-sounding speech than traditional neural text-to-speech (TTS) systems based on a cascade or two-stage approach. However, current E2E state-of-the-art models are computationally complex and memory-consuming, making them unsuitable for real-time offline on-device applications in low-resource scenarios. To address this issue, we propose a Lightweight E2E-TTS (LE2E) model that generates high-quality speech requiring minimal computational resources. We evaluate the proposed model on the LJSpeech dataset and show that it achieves state-of-the-art performance while being up to $90\%$ smaller in terms of model parameters and $10\times$ faster in real-time-factor. Furthermore, we demonstrate that the proposed E2E training paradigm achieves better quality compared to an equivalent architecture trained in a two-stage approach. Our results suggest that LE2E is a promising approach for developing real-time, high quality, low-resource TTS applications for on-device applications. |
| title | Lightweight End-to-end Text-to-speech Synthesis for low resource on-device applications |
| topic | Sound Artificial Intelligence Audio and Speech Processing |
| url | https://arxiv.org/abs/2505.07701 |