metadata
license: cc-by-4.0
language:
- ar
base_model:
- MBZUAI/artst_asr_v3_qasr
pipeline_tag: automatic-speech-recognition
import librosa
from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor
model_id="herwoww/nadi2025_task3_B1"
processor = AutoProcessor.from_pretrained(model_id)
model = AutoModelForSpeechSeq2Seq.from_pretrained(model_id).to(device)
pipe = pipeline(
"automatic-speech-recognition",
model=model,
tokenizer=processor.tokenizer,
feature_extractor=processor.feature_extractor,
torch_dtype=torch_dtype,
device=device,
)
wav, sr = librosa.load("audio.wav", sr=16000)
pipe(wav, generate_kwargs={'num_beams': 10, 'early_stopping': True})['text']