# Optimizing Neural Architectures for Hindi Speech Separation and Enhancement in Noisy Environments

arXiv: https://arxiv.org/abs/2508.12009 · HTML: https://mlsanity.com/paper/2508.12009/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 14 references that are in the map.
- First posted on arXiv 16 Aug 2025.

## Builds on (most cited first, 4 of 4)

- [Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference](https://mlsanity.com/paper/1712.05877/) (2017)
- [Quantizing deep convolutional networks for efficient inference: A whitepaper](https://mlsanity.com/paper/1806.08342/) (2018)
- [Supervised Speech Separation Based on Deep Learning: An Overview](https://mlsanity.com/paper/1708.07524/) (2017)
- [Demucs: Deep Extractor for Music Sources with extra unlabeled data remixed](https://mlsanity.com/paper/1909.01174/) (2019)

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

- [PhysicsSolutionAgent: Towards Multimodal Explanations for Numerical Physics Problem Solving](https://mlsanity.com/paper/2601.13453/) (2026)

## Abstract (arXiv)

This paper addresses the challenges of Hindi speech separation and enhancement using advanced neural network architectures, with a focus on edge devices. We propose a refined approach leveraging the DEMUCS model to overcome limitations of traditional methods, achieving substantial improvements in speech clarity and intelligibility. The model is fine-tuned with U-Net and LSTM layers, trained on a dataset of 400,000 Hindi speech clips augmented with ESC-50 and MS-SNSD for diverse acoustic environments. Evaluation using PESQ and STOI metrics shows superior performance, particularly under extreme noise conditions. To ensure deployment on resource-constrained devices like TWS earbuds, we explore quantization techniques to reduce computational requirements. This research highlights the effectiveness of customized AI algorithms for speech processing in Indian contexts and suggests future directions for optimizing edge-based architectures.