Spaces:
Sleeping
Sleeping
File size: 1,695 Bytes
1b4137e ed87615 56e1e3f 69b32e0 56e1e3f 69b32e0 1b4137e 56e1e3f 69b32e0 1b4137e 69b32e0 1b4137e 56e1e3f 1b4137e 56e1e3f 69b32e0 1b4137e 56e1e3f 1b4137e 69b32e0 1b4137e |
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 51 52 53 54 |
import gradio as gr
import subprocess
import os
from transformers import WhisperProcessor, WhisperForConditionalGeneration
import language_tool_python
from pydub import AudioSegment
from docx import Document
def extract_audio(video_path, audio_path):
command = f"ffmpeg -i '{video_path}' -ar 16000 -ac 1 -c:a pcm_s16le '{audio_path}' -y"
subprocess.run(command, shell=True, check=True)
return audio_path
def transcribe_audio(audio_path):
processor = WhisperProcessor.from_pretrained("openai/whisper-base")
model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-base")
audio_input = processor(audio_path, return_tensors="pt", sampling_rate=16000)
result = model.generate(**audio_input)
transcription = processor.decode(result[0], skip_special_tokens=True)
return transcription
def correct_text(text):
tool = language_tool_python.LanguageTool('es')
matches = tool.check(text)
return language_tool_python.utils.correct(text, matches)
def process_video(video_file):
video_path = video_file.name
audio_path = os.path.splitext(video_path)[0] + '.wav'
extract_audio(video_path, audio_path)
transcribed_text = transcribe_audio(audio_path)
corrected_text = correct_text(transcribed_text)
doc = Document()
doc.add_paragraph(corrected_text)
doc_path = "transcription.docx"
doc.save(doc_path)
return corrected_text, doc_path
demo = gr.Interface(
fn=process_video,
inputs=gr.File(label="Sube un archivo de video"),
outputs=[
gr.Textbox(label="Texto transcrito y corregido"),
gr.File(label="Descargar transcripción Word")
]
)
demo.launch()
|