# Correcting Mispronunciations in Speech using Spectrogram Inpainting

arXiv: https://arxiv.org/abs/2204.03379 · HTML: https://mlsanity.com/paper/2204.03379/

## 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 7 Apr 2022.

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

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis](https://mlsanity.com/paper/2010.05646/) (2020)
- [Image Inpainting for Irregular Holes Using Partial Convolutions](https://mlsanity.com/paper/1804.07723/) (2018)
- [WaveGlow: A Flow-based Generative Network for Speech Synthesis](https://mlsanity.com/paper/1811.00002/) (2018)
- [Vision-Infused Deep Audio Inpainting](https://mlsanity.com/paper/1910.10997/) (2019)
- [Parrotron: An End-to-End Speech-to-Speech Conversion Model and its Applications to Hearing-Impaired Speech and Speech Separation](https://mlsanity.com/paper/1904.04169/) (2019)
- [Inpainting of long audio segments with similarity graphs](https://mlsanity.com/paper/1607.06667/) (2016)
- [Deep speech inpainting of time-frequency masks](https://mlsanity.com/paper/1910.09058/) (2019)
- [Deep Long Audio Inpainting](https://mlsanity.com/paper/1911.06476/) (2019)
- [Replacing Human Audio with Synthetic Audio for On-device Unspoken Punctuation Prediction](https://mlsanity.com/paper/2010.10203/) (2020)
- [Introducing SPAIN (SParse Audio INpainter)](https://mlsanity.com/paper/1810.13137/) (2018)

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

- [Enhanced ASR Robustness to Packet Loss with a Front-End Adaptation Network](https://mlsanity.com/paper/2406.18928/) (2024)

## Abstract (arXiv)

Learning a new language involves constantly comparing speech productions with reference productions from the environment. Early in speech acquisition, children make articulatory adjustments to match their caregivers' speech. Grownup learners of a language tweak their speech to match the tutor reference. This paper proposes a method to synthetically generate correct pronunciation feedback given incorrect production. Furthermore, our aim is to generate the corrected production while maintaining the speaker's original voice. The system prompts the user to pronounce a phrase. The speech is recorded, and the samples associated with the inaccurate phoneme are masked with zeros. This waveform serves as an input to a speech generator, implemented as a deep learning inpainting system with a U-net architecture, and trained to output a reconstructed speech. The training set is composed of unimpaired proper speech examples, and the generator is trained to reconstruct the original proper speech. We evaluated the performance of our system on phoneme replacement of minimal pair words of English as well as on children with pronunciation disorders. Results suggest that human listeners slightly prefer our generated speech over a smoothed replacement of the inaccurate phoneme with a production of a different speaker.