# Low-resource ASR

SpeechLLM is a multimodal large language model designed to predict speaker turn metadata in conversations, including speech activity, transcribed text, gender, age, accent, and emotion.

A Luxembourgish speech recognition model fine-tuned from the wav2vec 2.0 large XLSR-53 checkpoint, trained with 842 hours of unlabeled and 14 hours of labeled data, integrated with a 5-gram language model

A Hindi speech recognition model fine-tuned from facebook/wav2vec2-large-xlsr-53, trained on multilingual and code-switching ASR challenge data for low-resource Indian languages

This is an automatic speech recognition (ASR) model fine-tuned on the MOZILLA-FOUNDATION/COMMON_VOICE_7_0 - AB dataset, based on the XLS-R architecture