metadata
license: apache-2.0
language:
- new
datasets:
- ilprl-docse/Nwacha_Muna_A_Newari_ASR_Dataset
metrics:
- cer
- wer
pipeline_tag: automatic-speech-recognition
tags:
- asr
- nepal-bhasha
- newari
- conformer
- low-resource
- semi-supervised
Nwāchā Munā NepConformer Semi-Supervised Learning (SSL) Approach
This model was trained as part of the paper Nwāchā Munā: A Devanagari Speech Corpus and Proximal Transfer Benchmark for Nepal Bhasha ASR. It is a NepConformer fine-tuned on the Nwāchā Munā corpus with a semi-supervised learning approach — achieving 19.83% CER. The training scripts can be found at github.com/ilprl/nwacha-muna.
Usage
import nemo.collections.asr as nemo_asr
model = nemo_asr.models.ASRModel.from_pretrained("ilprl-docse/NwachaMuna-NepConformer-SSL")
transcriptions = model.transcribe(["file.wav"])
print(transcriptions[0])