# Mispronunciation Detection and Diagnosis Without Model Training: A Retrieval-Based Approach

arXiv: https://arxiv.org/abs/2511.20107 · HTML: https://mlsanity.com/paper/2511.20107/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 21 references that are in the map.
- First posted on arXiv 25 Nov 2025.

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

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://mlsanity.com/paper/2106.07447/) (2021)
- [Retrieval-Augmented Generation for Large Language Models: A Survey](https://mlsanity.com/paper/2312.10997/) (2023)
- [data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language](https://mlsanity.com/paper/2202.03555/) (2022)
- [Voice Conversion With Just Nearest Neighbors](https://mlsanity.com/paper/2305.18975/) (2023)
- [A Full Text-Dependent End to End Mispronunciation Detection and Diagnosis with Easy Data Augmentation Techniques](https://mlsanity.com/paper/2104.08428/) (2021)
- [An Approach to Mispronunciation Detection and Diagnosis with Acoustic, Phonetic and Linguistic (APL) Embeddings](https://mlsanity.com/paper/2110.07274/) (2021)
- [Multi-View Multi-Task Representation Learning for Mispronunciation Detection](https://mlsanity.com/paper/2306.01845/) (2023)
- [Text-Aware End-to-end Mispronunciation Detection and Diagnosis](https://mlsanity.com/paper/2206.07289/) (2022)
- [L1-aware Multilingual Mispronunciation Detection Framework](https://mlsanity.com/paper/2309.07719/) (2023)

## Abstract (arXiv)

Mispronunciation Detection and Diagnosis (MDD) is crucial for language learning and speech therapy. Unlike conventional methods that require scoring models or training phoneme-level models, we propose a novel training-free framework that leverages retrieval techniques with a pretrained Automatic Speech Recognition model. Our method avoids phoneme-specific modeling or additional task-specific training, while still achieving accurate detection and diagnosis of pronunciation errors. Experiments on the L2-ARCTIC dataset show that our method achieves a superior F1 score of 69.60% while avoiding the complexity of model training.