File size: 1,527 Bytes
670f5ce |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 |
import gradio as gr
import torch
import torchaudio
import numpy as np
from transformers import Wav2Vec2BertProcessor, Wav2Vec2BertForCTC
# Load model and processor
repo_id = "hriteshMaikap/marathi-asr-model"
processor = Wav2Vec2BertProcessor.from_pretrained(repo_id)
model = Wav2Vec2BertForCTC.from_pretrained(repo_id)
device = "cuda" if torch.cuda.is_available() else "cpu"
model = model.to(device)
def transcribe(audio):
# Process audio
waveform, sample_rate = torchaudio.load(audio)
# Resample if needed
if sample_rate != 16000:
resampler = torchaudio.transforms.Resample(sample_rate, 16000)
waveform = resampler(waveform)
# Convert to mono if needed
if waveform.shape[0] > 1:
waveform = torch.mean(waveform, dim=0, keepdim=True)
# Convert to numpy
speech_array = waveform.squeeze().numpy()
# Process and run inference
with torch.no_grad():
inputs = processor(speech_array, sampling_rate=16000, return_tensors="pt").to(device)
logits = model(inputs.input_features).logits
predicted_ids = torch.argmax(logits, dim=-1)
# Decode the predicted IDs
transcription = processor.decode(predicted_ids[0])
return transcription
# Create Gradio interface
iface = gr.Interface(
fn=transcribe,
inputs=gr.Audio(source="microphone", type="filepath"),
outputs="text",
title="Marathi Speech Recognition",
description="Record your voice in Marathi and get a transcription."
)
iface.launch() |