abidanoaman/whisper-tiny-fast-urdu-finetuned-common-voice-partaial

from transformers import WhisperProcessor, WhisperForConditionalGeneration
import torchaudio

processor = WhisperProcessor.from_pretrained("abidanoaman/whisper-tiny-fast-urdu-finetuned-common-voice-partaial")
model     = WhisperForConditionalGeneration.from_pretrained("abidanoaman/whisper-tiny-fast-urdu-finetuned-common-voice-partaial")

audio, sr = torchaudio.load("audio.wav")
if sr != 16000:
    audio = torchaudio.transforms.Resample(sr, 16000)(audio)
audio = audio.squeeze().numpy()

inputs = processor(audio, sampling_rate=16000, return_tensors="pt").input_features
predicted_ids = model.generate(inputs)
transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)[0]
print(transcription)

from huggingface_hub import snapshot_download
from faster_whisper import WhisperModel

# Download faster-whisper model
model_path = snapshot_download(
    repo_id="abidanoaman/whisper-tiny-fast-urdu-finetuned-common-voice-partaial",
    allow_patterns="faster-whisper/*"
)

model = WhisperModel(
    f"{model_path}/faster-whisper",
    device="cpu",
    compute_type="int8"
)

segments, info = model.transcribe(
    "audio.wav",
    language="ur",
    vad_filter=True,
    beam_size=5,
    condition_on_previous_text=False,
    temperature=0.0,
)

text = " ".join([seg.text for seg in segments])
print(text)