Audio Datasets Mine
Open sourced audio datasets for STT/ASR. All recordings by me (Daniel Rosehill) unless otherwise accredited.
About This Project
This project hosts a repository of curated audio datasets on Hugging Face, primarily featuring recordings produced by Daniel Rosehill. The datasets are designed to provide high-quality, labeled audio data for researchers and developers working on speech-to-text (STT) and automatic speech recognition (ASR) model development.
The collection serves as a resource for those looking to fine-tune or evaluate machine learning models for speech recognition. By providing accessible, self-recorded audio data, it supports the development of more robust voice processing applications and contributes to the open-source data ecosystem for natural language processing.