Parakeet TDT 0.6B V2

Automatic speech recognition model using the Transducer architecture. It is built with the NeMo library and supports English language. The model is trained on datasets such as nvidia/Granary and nvidia/nemo-asr-set-3.0. It features tags like FastConformer, Conformer, and is part of the hf-asr-leaderboard. Example audio samples include Librispeech sample 1 and Librispeech sample 2.