Downloads · 30 days
26
40% of all-time downloads
G999n/whisper-large-v3-turbo-odia-lora
whisper-large-v3-turbo-odia-lora is a automatic speech recognition model from G999n. Use it when you need speech turned into text. It is set up for peft. The card lists the license as apache-2.0.
This repository contains the fine-tuned LoRA (PEFT) adapter weights for openai/whisper-large-v3-turbo on Odia speech from the ai4bharat/IndicVoices corpus.
Downloads · 30 days
26
40% of all-time downloads
All-time downloads
65
Public
Repo size
105 MB
Likes
0
Public
Click a slice to open those files.
.safetensors52.5 MB · 93%
From the Hugging Face model README
This repository contains the fine-tuned LoRA (PEFT) adapter weights for openai/whisper-large-v3-turbo on Odia speech from the ai4bharat/IndicVoices corpus.
openai/whisper-large-v3-turboor / proxy script guidance bengali)ai4bharat/IndicVoices (Odia split)64128q_proj, v_proj11.8M / 809M (1.46%)import torch
import librosa
from transformers import WhisperProcessor, WhisperForConditionalGeneration
from peft import PeftModel
base_model_id = "openai/whisper-large-v3-turbo"
peft_model_id = "G999n/whisper-large-v3-turbo-odia-lora"
# 1. Load processor & base model
processor = WhisperProcessor.from_pretrained(base_model_id, language="bengali", task="transcribe")
model = WhisperForConditionalGeneration.from_pretrained(base_model_id, torch_dtype=torch.float16, device_map="auto")
# 2. Attach LoRA adapter
model = PeftModel.from_pretrained(model, peft_model_id)
model.eval()
# 3. Transcribe audio
audio, sr = librosa.load("your_odia_audio.wav", sr=16000)
input_features = processor(audio, sampling_rate=16000, return_tensors="pt").input_features.to("cuda", dtype=torch.float16)
with torch.no_grad():
predicted_ids = model.generate(input_features, max_length=225)
transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)[0]
print("Transcribed Odia Text:", transcription)
Trained on Google Cloud Platform using an NVIDIA L4 GPU.