# CUPE: Contextless Universal Phoneme Encoder for Language-Agnostic Speech Processing

arXiv: https://arxiv.org/abs/2508.15316 · HTML: https://mlsanity.com/paper/2508.15316/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 18 references that are in the map.
- First posted on arXiv 21 Aug 2025.

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

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Squeeze-and-Excitation Networks](https://mlsanity.com/paper/1709.01507/) (2017)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale](https://mlsanity.com/paper/2111.09296/) (2021)
- [Unsupervised Cross-lingual Representation Learning for Speech Recognition](https://mlsanity.com/paper/2006.13979/) (2020)
- [FLEURS: Few-shot Learning Evaluation of Universal Representations of Speech](https://mlsanity.com/paper/2205.12446/) (2022)
- [ContextNet: Improving Convolutional Neural Networks for Automatic Speech Recognition with Global Context](https://mlsanity.com/paper/2005.03191/) (2020)
- [Very Deep Convolutional Neural Networks for Raw Waveforms](https://mlsanity.com/paper/1610.00087/) (2016)
- [Multistream CNN for Robust Acoustic Modeling](https://mlsanity.com/paper/2005.10470/) (2020)
- [Allophant: Cross-lingual Phoneme Recognition with Articulatory Attributes](https://mlsanity.com/paper/2306.04306/) (2023)
- [Hierarchical Softmax for End-to-End Low-resource Multilingual Speech Recognition](https://mlsanity.com/paper/2204.03855/) (2022)

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

- [Extracting accent features in spoken Brazilian Portuguese without sociolinguistic labels](https://mlsanity.com/paper/2605.30457/) (2026)

## Abstract (arXiv)

Universal phoneme recognition typically requires analyzing long speech segments and language-specific patterns. Many speech processing tasks require pure phoneme representations free from contextual influence, which motivated our development of CUPE - a lightweight model that captures key phoneme features in just 120 milliseconds, about one phoneme's length. CUPE processes short, fixed-width windows independently and, despite fewer parameters than current approaches, achieves competitive cross-lingual performance by learning fundamental acoustic patterns common to all languages. Our extensive evaluation through supervised and self-supervised training on diverse languages, including zero-shot tests on the UCLA Phonetic Corpus, demonstrates strong cross-lingual generalization and reveals that effective universal speech processing is possible through modeling basic acoustic patterns within phoneme-length windows.