Are you sure you want to delete this access key?
Legend |
---|
DVC Managed File |
Git Managed File |
Metric |
Stage File |
External File |
Legend |
---|
DVC Managed File |
Git Managed File |
Metric |
Stage File |
External File |
Automatic Speech Recognition using Facebook wav2vec2-xls-r-300m model and mozilla-foundation common_voice_8_0 Urdu Dataset.
This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice dataset.
It achieves the following results on the evaluation set:
To evaluate on mozilla-foundation/common_voice_8_0
with split test
python3 ./eval.py --model_id ./Model --dataset ./Data --config ur --split test --chunk_length_s 5.0 --stride_length_s 1.0 --log_outputs
import torch
from datasets import load_dataset, Audio
from transformers import pipeline
import torchaudio.functional as F
model = "Model"
data = load_dataset("Data", "ur", split="test", delimiter="\t")
def path_adjust(batch):
batch["path"] = "Data/ur/clips/" + str(batch["path"])
return batch
data = data.map(path_adjust)
sample_iter = iter(data.cast_column("path", Audio(sampling_rate=16_000)))
sample = next(sample_iter)
asr = pipeline("automatic-speech-recognition", model=model)
prediction = asr(
sample["path"]["array"], chunk_length_s=5, stride_length_s=1)
prediction
# => {'text': 'اب یہ ونگین لمحاتانکھار دلمیں میںفوث کریلیا اجائ'}
Without LM | With LM (run ./eval.py ) |
---|---|
56.21 | 46.37 |
Press p or to see the previous file or, n or to see the next file
Automatic Speech Recognition using Facebook wav2vec2-xls-r-300m model and mozilla-foundation common_voice_8_0 Urdu Dataset
https://huggingface.co/kingabzpro/wav2vec2-large-xls-r-300m-UrduAre you sure you want to delete this access key?
Are you sure you want to delete this access key?
Are you sure you want to delete this access key?
Are you sure you want to delete this access key?
commented in commite2f5996ee7on branch master
2 years agoI am open to contributions and suggestion. So, Keep them comming.
commented in commite2f5996ee7
1 year agoI got 17% WER and I am still far from SOTA.