nvidia

/

parakeet-rnnt-110m-da-dk

like 15

Follow

NVIDIA 46.7k

Hey, this project looks super cool! Thanks for taking the time and effort to train a danish ASR model.

I wanted to make a small Gradio demo app for the danish community to try out the model in a HuggingFace Space. However, I'm having some installation issues.

After installing nemo_toolkit[asr] with Python 3.10.12 I follow the instructions:

➜  parakeet-da-app uv run python   
Python 3.10.12 (main, Jul 26 2023, 19:37:41) [Clang 16.0.3 ] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import nemo.collections.asr as nemo_asr
>>> asr_model = nemo_asr.models.ASRModel.from_pretrained(model_name="nvidia/parakeet-rnnt-110m-da-dk")
[NeMo I 2025-10-18 16:07:07 common:964] Found 1 files in cache directory /Users/rasgaard/.cache/torch/NeMo/NeMo_2.5.0/hf_hub_cache/nvidia/parakeet-rnnt-110m-da-dk
[NeMo I 2025-10-18 16:07:07 common:965] Deleting old cache directory for model `nvidia/parakeet-rnnt-110m-da-dk` in order to prevent duplicates...
[NeMo I 2025-10-18 16:07:07 common:971] Downloading nvidia/parakeet-rnnt-110m-da-dk from HuggingFace Hub to path: /Users/rasgaard/.cache/torch/NeMo/NeMo_2.5.0/hf_hub_cache/nvidia/parakeet-rnnt-110m-da-dk/fe9901f674f2ecb0bd625af577bf5524
.gitattributes: 1.56kB [00:00, 4.28MB/s]                     | 0/3 [00:00<?, ?it/s]
README.md: 6.86kB [00:00, 24.3MB/s]                  | 1/3 [00:00<00:00,  3.68it/s]
stt_da_fastconformer_rnnt_large.nemo: 100%|██████| 452M/452M [00:00<00:00, 553MB/s]
Fetching 3 files: 100%|██████████████████████████████| 3/3 [00:01<00:00,  2.61it/s]
[NeMo I 2025-10-18 16:07:09 save_restore_connector:142] Restoration will occur within pre-extracted directory : `/Users/rasgaard/.cache/torch/NeMo/NeMo_2.5.0/hf_hub_cache/nvidia/parakeet-rnnt-110m-da-dk/fe9901f674f2ecb0bd625af577bf5524`.
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/Users/rasgaard/repos/parakeet-da-app/.venv/lib/python3.10/site-packages/nemo/core/classes/common.py", line 822, in from_pretrained
    instance = class_.restore_from(
  File "/Users/rasgaard/repos/parakeet-da-app/.venv/lib/python3.10/site-packages/nemo/core/classes/modelPT.py", line 493, in restore_from
    instance = cls._save_restore_connector.restore_from(
  File "/Users/rasgaard/repos/parakeet-da-app/.venv/lib/python3.10/site-packages/nemo/core/connectors/save_restore_connector.py", line 267, in restore_from
    loaded_params = self.load_config_and_state_dict(
  File "/Users/rasgaard/repos/parakeet-da-app/.venv/lib/python3.10/site-packages/nemo/core/connectors/save_restore_connector.py", line 165, in load_config_and_state_dict
    conf = OmegaConf.load(config_yaml)
  File "/Users/rasgaard/repos/parakeet-da-app/.venv/lib/python3.10/site-packages/omegaconf/omegaconf.py", line 189, in load
    with io.open(os.path.abspath(file_), "r", encoding="utf-8") as f:
FileNotFoundError: [Errno 2] No such file or directory: '/Users/rasgaard/.cache/torch/NeMo/NeMo_2.5.0/hf_hub_cache/nvidia/parakeet-rnnt-110m-da-dk/fe9901f674f2ecb0bd625af577bf5524/model_config.yaml'