# Towards Stress-Aware Sentence-Level Filipino G2P With Weakly-Supervised ByT5 Fine-Tuning

arXiv: https://arxiv.org/abs/2609.09974 · HTML: https://mlsanity.com/paper/2609.09974/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 27 references that are in the map.
- First posted on arXiv 9 Sep 2026.

## Builds on (most cited first, 11 of 11)

- [Conformer: Convolution-augmented Transformer for Speech Recognition](https://mlsanity.com/paper/2005.08100/) (2020)
- [mT5: A massively multilingual pre-trained text-to-text transformer](https://mlsanity.com/paper/2010.11934/) (2020)
- [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://mlsanity.com/paper/2105.13626/) (2021)
- [ByT5 model for massively multilingual grapheme-to-phoneme conversion](https://mlsanity.com/paper/2204.03067/) (2022)
- [Transformer based Grapheme-to-Phoneme Conversion](https://mlsanity.com/paper/2004.06338/) (2020)
- [Sequence-to-Sequence Neural Net Models for Grapheme-to-Phoneme Conversion](https://mlsanity.com/paper/1506.00196/) (2015)
- [Exploiting News Article Structure for Automatic Corpus Generation of Entailment Datasets](https://mlsanity.com/paper/2010.11574/) (2020)
- [One Model to Pronounce Them All: Multilingual Grapheme-to-Phoneme Conversion With a Transformer Ensemble](https://mlsanity.com/paper/2006.13343/) (2020)
- [Transcribing Bengali Text with Regional Dialects to IPA using District Guided Tokens](https://mlsanity.com/paper/2403.17407/) (2024)
- [Fast, Not Fancy: Rethinking G2P with Rich Data and Rule-Based Models](https://mlsanity.com/paper/2505.12973/) (2025)
- [Accent Placement Models for Rigvedic Sanskrit Text](https://mlsanity.com/paper/2511.23088/) (2025)

## Built on by (most cited first, 1 of 1)

- [TDR: Tagalog Diacritic Restoration Dataset](https://mlsanity.com/paper/2610.17597/) (2026)

## Abstract (arXiv)

Grapheme-to-phoneme conversion (G2P) refers to the task of converting a sequence of graphemes to a corresponding sequence of phonemes. While Filipino G2P is fairly straightforward due to its shallow orthography, the inclusion of prosodic features such as stress adds a layer of complexity that requires sentence-level context instead of single-word inputs. However, sentence-level data for Filipino typically do not include phoneme transcriptions, posing a challenge for training G2P models. As such, we investigate how to obtain sentence-level phoneme data for Filipino using available data and compare the resulting models with multilingual word-level G2P as well as measure how accurately they predict stress marker position for Filipino. We propose fine-tuning a ByT5-based model, pre-trained on multilingual word-level G2P data, on three sentence-level G2P datasets annotated with an LLM-assisted pipeline guided by data from Wiktionary. This approach produces models that perform well on the G2P task, achieving at best around 0.54% PER and 2.50% CER, a significant decrease compared to base model PER at around 19.74%, on a manually-corrected test set. The model is able to correctly classify most of the main stress classes in Filipino, but struggles particularly with malumi words. We show that a ByT5-based model performs well at sentence-level Filipino G2P and offers strong potential for Filipino homograph disambiguation.