Spaces:
Sleeping
Sleeping
File size: 1,564 Bytes
88dc3ba 5d9c950 b2593bc 5d9c950 527e644 b2593bc f47a9e0 78cc121 b2593bc f47a9e0 b2593bc 038e82c 7d6796c 52fc07d 78cc121 52fc07d 038e82c 52fc07d 038e82c 52fc07d 038e82c 52fc07d 527e644 b2593bc 038e82c |
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 |
import subprocess
subprocess.run(["pip", "install", "gradio", "--upgrade"])
subprocess.run(["pip", "install", "transformers"])
subprocess.run(["pip", "install", "torchaudio", "--upgrade"])
import gradio as gr
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
import torchaudio
# Load model and processor
processor = Wav2Vec2Processor.from_pretrained("jonatasgrosman/wav2vec2-large-xlsr-53-italian")
model = Wav2Vec2ForCTC.from_pretrained("jonatasgrosman/wav2vec2-large-xlsr-53-italian")
# Function to perform ASR on audio data
def transcribe_audio(audio_data):
print("Received audio data:", audio_data) # Debug print
if audio_data is None:
return "No audio data received."
try:
# Convert audio data to mono and normalize
audio_data = torchaudio.transforms.Resample(audio_data[1], 16000)(audio_data[0])
audio_data = torchaudio.functional.gain(audio_data, gain_db=5.0)
# Apply custom preprocessing to the audio data if needed
input_values = processor(audio_data[0].numpy(), return_tensors="pt").input_values
# Perform ASR
with torch.no_grad():
logits = model(input_values).logits
# Decode the output
predicted_ids = torch.argmax(logits, dim=-1)
transcription = processor.batch_decode(predicted_ids)
return transcription[0]
except Exception as e:
return f"An error occurred: {str(e)}"
# Create Gradio interface
audio_input = gr.Audio()
gr.Interface(fn=transcribe_audio, inputs=audio_input, outputs="text").launch()
|