Official documentation
AuroraAI-Research/AuroraASR-1M · Hugging Face
AuroraAI Research, Hugging Face (model card by AuroraAI Research), 2026
Read the originalRead on October 23, 2026
- What it is
- The model card for AuroraASR-1M, an automatic speech recognition model that AuroraAI Research publishes on Hugging Face under the Apache 2.0 license. It says the model was trained on the full LibriSpeech dataset in English only, and it reports word error rates on the English LibriSpeech test sets.
- How much weight it carries
- The maker's own model card, read directly on 2026-10-23 (static HTML). It is the maker's documentation, so it is the authority on the training language. It lists no set of supported languages, so the count of one language is drawn from its training statement. The benchmark figures are the maker's own and were not checked independently.
Facts taken from this source
- AuroraAI Research's model card says AuroraASR-1M was trained on the full LibriSpeech dataset in English only, so the model's count is one language.
“Trained on the full LibriSpeech dataset for 379 epochs, English only, 16 kHz mono input” (Model card, training description)