Create app.py

app.py

@@ -0,0 +1,58 @@
+import gradio as gr
+import torchaudio
+import torch
+import librosa
+import numpy as np
+import moviepy.editor as mp
+from transformers import Wav2Vec2Processor, Wav2Vec2ForCTC
+
+model_name = "m3hrdadfi/wav2vec2-large-xlsr-persian"
+processor = Wav2Vec2Processor.from_pretrained(model_name)
+model = Wav2Vec2ForCTC.from_pretrained(model_name)
+
+def preprocess_audio(audio_path):
+    y, sr = librosa.load(audio_path, sr=16000, mono=True) 
+    y = librosa.util.normalize(y)  
+    y = librosa.effects.preemphasis(y) 
+    return torch.tensor(y, dtype=torch.float32).unsqueeze(0)  
+
+
+def speech_to_text(audio_path):
+    waveform = preprocess_audio(audio_path)
+    
+    input_values = processor(waveform.squeeze(), return_tensors="pt", sampling_rate=16000).input_values
+    with torch.no_grad():
+        logits = model(input_values).logits
+    predicted_ids = torch.argmax(logits, dim=-1)
+    transcription = processor.batch_decode(predicted_ids)[0]
+    return transcription
+
+
+def video_to_text(video_path):
+    video = mp.VideoFileClip(video_path)
+    audio_path = "extracted_audio.wav"
+    video.audio.write_audiofile(audio_path, codec="pcm_s16le")
+    return speech_to_text(audio_path)
+
+with gr.Blocks() as demo:
+    gr.Markdown("### تبدیل گفتار فارسی به متن با استفاده از Wav2Vec2")
+
+    with gr.Tab("آپلود ویدئو"):
+        video_input = gr.File(label="انتخاب ویدئو")
+        video_output = gr.Textbox(label="متن استخراج شده")
+        video_button = gr.Button("تبدیل به متن")
+        video_button.click(video_to_text, inputs=video_input, outputs=video_output)
+
+    with gr.Tab("آپلود فایل صوتی"):
+        audio_input = gr.File(label="انتخاب فایل صوتی")
+        audio_output = gr.Textbox(label="متن استخراج شده")
+        audio_button = gr.Button("تبدیل به متن")
+        audio_button.click(speech_to_text, inputs=audio_input, outputs=audio_output)
+
+    with gr.Tab("ضبط صدا"):
+        mic_input = gr.Audio(sources="microphone", type="filepath")
+        mic_output = gr.Textbox(label="متن استخراج شده")
+        mic_button = gr.Button("تبدیل به متن")
+        mic_button.click(speech_to_text, inputs=mic_input, outputs=mic_output)
+
+demo.launch()