This is a fine-tuned version of the XLS-R wav2vec2 large model (53 languages) further specialized for Amharic automatic speech recognition. Built with the Hugging Face Transformers library, it enables developers to perform ASR on Amharic speech data with improved accuracy over the base multilingual model.
In the Speech to text space, Wav2vec2 Large Xlsr 53 Amharic takes a focused approach. It focuses on transcribing spoken Amharic audio into text using open-source speech models. Wav2vec2 Large Xlsr 53 Amharic is an open-source project aimed at developers. The project is open source (Open Source). Wav2vec2 Large Xlsr 53 Amharic is available on the web and API.
agkphysics builds and maintains Wav2vec2 Large Xlsr 53 Amharic, and it first shipped in 2023.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do