Update requirements.txt

requirements.txt

@@ -1,7 +1,60 @@
-gradio
-whisper
-spacy
-language_tool_python
-python-docx
-moviepy
-imageio[ffmpeg]
+import os
+import subprocess
+
+# Asegurar instalación de moviepy y ffmpeg
+subprocess.run(["pip", "install", "moviepy", "imageio[ffmpeg]", "imageio_ffmpeg"], check=True)
+
+import whisper
+import spacy
+import language_tool_python
+import gradio as gr
+from moviepy.editor import VideoFileClip
+from docx import Document
+
+def extract_audio(video_path, audio_path):
+    try:
+        video = VideoFileClip(video_path)
+        audio = video.audio
+        audio.write_audiofile(audio_path, codec='pcm_s16le', fps=16000)
+        return True
+    except Exception as e:
+        print(f"Error al extraer audio: {e}")
+        return False
+
+def transcribe_audio(audio_path):
+    model = whisper.load_model("base")
+    result = model.transcribe(audio_path, word_timestamps=True)
+    return result
+
+def correct_text(text):
+    tool = language_tool_python.LanguageTool('es')
+    matches = tool.check(text)
+    return language_tool_python.utils.correct(text, matches)
+
+def create_word_doc(segments, output_path):
+    doc = Document()
+    for segment in segments:
+        corrected_text = correct_text(segment['text'])
+        doc.add_paragraph(corrected_text)
+    doc.save(output_path)
+    return output_path
+
+def process_video(video_file):
+    audio_path = video_file.replace(".mp4", ".wav")
+    word_output = video_file.replace(".mp4", "_transcription.docx")
+    
+    if extract_audio(video_file, audio_path):
+        result = transcribe_audio(audio_path)
+        segments = result['segments']
+        doc_path = create_word_doc(segments, word_output)
+        return "Transcripción completada.", doc_path
+    else:
+        return "Error al procesar el archivo.", None
+
+demo = gr.Interface(
+    fn=process_video,
+    inputs=gr.File(label="Sube un archivo de video"),
+    outputs=["text", gr.File(label="Descargar transcripción")]
+)
+
+demo.launch()