Automatic Speech Recognition
NeMo
Safetensors
PyTorch
sortformer
speaker-diarization
speaker-recognition
speech
audio
Transformer
FastConformer
Conformer
NEST
NeMo
Eval Results (legacy)
Instructions to use nvidia/diar_sortformer_4spk-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- NeMo
How to use nvidia/diar_sortformer_4spk-v1 with NeMo:
import nemo.collections.asr as nemo_asr asr_model = nemo_asr.models.ASRModel.from_pretrained("nvidia/diar_sortformer_4spk-v1") transcriptions = asr_model.transcribe(["file.wav"]) - Notebooks
- Google Colab
- Kaggle
model weights of canary-1b with adapter
#6
by javiabellan - opened
i am interested in doing the full multispeaker ASR. The paper mentions both a fintuned canary-170M and canary-1B with adapters. Both model can do multispeaker ASR, they will be open source?
same question
Unfortunately, we have no plans on releasing offline multi-speaker Canary model. You can still implement multispeaker Canary using NeMo toolkit by following the paper.
Thanks for answering. Sad to hear that.
javiabellan changed discussion status to closed