Automatic Speech Recognition
NeMo
Safetensors
PyTorch
English
speech
audio
Transformer
FastConformer
Conformer
NeMo
Qwen
hf-asr-leaderboard
Eval Results (legacy)
Eval Results
Instructions to use nvidia/canary-qwen-2.5b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- NeMo
How to use nvidia/canary-qwen-2.5b with NeMo:
import nemo.collections.asr as nemo_asr asr_model = nemo_asr.models.ASRModel.from_pretrained("nvidia/canary-qwen-2.5b") transcriptions = asr_model.transcribe(["file.wav"]) - Notebooks
- Google Colab
- Kaggle
NeMo container for canary-qwen-2.5b
#3
by xinyuenn - opened
Can you directly provide a Nemo container that supports canary-qwen-2.5b model inference? Or provide the version numbers of the main libraries, such as torch, nemo, megatron?
You'll need:
- PyTorch 2.6.0+
- the latest unreleased NeMo (2.5.0 release will have all the necessary code): use
python -m pip install "nemo_toolkit[asr,tts] @ git+https://github.com/NVIDIA/NeMo.git"
Megatron is not needed. Generally it should work OK in any venv/conda env/etc with the two items above installed.