NVIDIA Parakeet 110M

NVIDIA Parakeet 110M is a highly efficient, hybrid automatic speech recognition model designed for fast and accurate English transcription with automatic punctuation and capitalization.

Developed jointly by the NVIDIA NeMo and Suno.ai teams, NVIDIA Parakeet 110M (specifically parakeet-tdt_ctc-110m) is a 114-million-parameter automatic speech recognition model built on a hybrid FastConformer architecture. By combining Token-and-Duration Transducer (TDT) and Connectionist Temporal Classification (CTC) decoders, this model achieves faster training convergence and highly efficient inference. It delivers robust English text transcription, complete with accurate capitalization and punctuation, making it an ideal lightweight choice for real-time speech-to-text pipelines.

https://catalog.ngc.nvidia.com/orgs/nvidia/teams/nemo/models/parakeet-tdt_ctc-110m