app.py

import gradio as gr
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
import torch
import torchaudio

# Load the pre-trained Wav2Vec 2.0 model and processor from Hugging Face
processor = Wav2Vec2Processor.from_pretrained("facebook/wav2vec2-large-960h")
model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-large-960h")

# Function to convert speech to text
def speech_to_text(audio_file):
    # Load the audio file
    audio_input, _ = torchaudio.load(audio_file)
    
    # Preprocess the audio input (e.g., resample, normalize, etc.)
    input_values = processor(audio_input, return_tensors="pt").input_values
    
    # Perform speech-to-text (CTC Decoding)
    with torch.no_grad():
        logits = model(input_values).logits
    predicted_ids = torch.argmax(logits, dim=-1)

    # Decode the predicted ids to text
    transcription = processor.decode(predicted_ids[0])
    
    return transcription

# Set up the Gradio interface
iface = gr.Interface(
    fn=speech_to_text,  # Function to be executed
    inputs=gr.Audio(type="filepath"),  # Correct type for file upload
    outputs=gr.Textbox(),  # Display transcription in a text box
    title="Speech-to-Text Analyzer for Lectimport gradio as gr
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
import torch
import torchaudio

# Load the pre-trained Wav2Vec 2.0 model and processor from Hugging Face
processor = Wav2Vec2Processor.from_pretrained("facebook/wav2vec2-large-960h")
model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-large-960h")

# Function to convert speech to text
def speech_to_text(audio_file):
    # Load the audio file
    audio_input, _ = torchaudio.load(audio_file)
    
    # Preprocess the audio input (e.g., resample, normalize, etc.)
    input_values = processor(audio_input, return_tensors="pt").input_values
    
    # Perform speech-to-text (CTC Decoding)
    with torch.no_grad():
        logits = model(input_values).logits
    predicted_ids = torch.argmax(logits, dim=-1)

    # Decode the predicted ids to text
    transcription = processor.decode(predicted_ids[0])
    
    return transcription

# Set up the Gradio interface
iface = gr.Interface(
    fn=speech_to_text,  # Function to be executed
    inputs=gr.Audio(type="filepath"),  # Correct type for file upload
    outputs=gr.Textbox(),  # Display transcription in a text box
    title="Speech-to-Text Analyzer for Lecture Notes",
    description="Upload an audio file (e.g., lecture recording) to get the transcription of the speech."
)

# Launch the interface
iface.launch()
ure Notes",
    description="Upload an audio file (e.g., lecture recording) to get the transcription of the speech."
)

# Launch the interface
iface.launch()