Spaces:
Sleeping
Sleeping
File size: 2,264 Bytes
966d998 35e43fd 8107aa0 e3bb042 35e43fd 3bfabc7 35e43fd 3bfabc7 35e43fd 3bfabc7 8107aa0 3bfabc7 35e43fd 966d998 35e43fd 966d998 8107aa0 966d998 3bfabc7 966d998 3bfabc7 966d998 711a612 55884fe cbaf479 | 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 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 | import gradio as gr
import yt_dlp
import torch
import numpy as np
from faster_whisper import WhisperModel
import os
os.system("pip install -q gradio torch whisper faster-whisper yt-dlp numpy")
# Load model sekali aja
model = WhisperModel("small", device="cpu", compute_type="float32")
# Fungsi untuk transkripsi dari file
def transcribe_audio(file):
segments, _ = model.transcribe(file)
transcript = "\n".join(segment.text for segment in segments)
# Simpan ke file TXT
file_path = "transcription.txt"
with open(file_path, "w", encoding="utf-8") as f:
f.write(transcript)
return f"**Transcription:**\n{transcript}"
# Fungsi untuk ambil audio dari YouTube
def get_audio_from_youtube(url):
ydl_opts = {
"format": "bestaudio/best",
"postprocessors": [{"key": "FFmpegExtractAudio", "preferredcodec": "mp3", "preferredquality": "192"}],
"outtmpl": "temp_audio.%(ext)s",
}
with yt_dlp.YoutubeDL(ydl_opts) as ydl:
info = ydl.extract_info(url, download=True)
return "temp_audio.mp3"
# Fungsi untuk transkripsi dari YouTube
def transcribe_youtube(url):
audio_file = get_audio_from_youtube(url)
return transcribe_audio(audio_file)
#memastikan file ada sebelum digunakan
if not os.path.exists('transcription.txt'):
with open("transcription.txt", "w", encoding="utf-8") as f:
f.write("")
# Gradio UI
with gr.Blocks() as app:
gr.Markdown("# 🎤 YouTube & Audio Transcriber with Whisper AI")
with gr.Tab("Upload File"):
audio_input = gr.File(label="Upload Audio File")
file_transcribe_button = gr.Button("Transcribe")
file_output = gr.Textbox(label="Transcription")
download_file = gr.File(label="Download Transcription", value="transcription.txt")
with gr.Tab("YouTube Video"):
youtube_url = gr.Textbox(label="YouTube URL")
yt_transcribe_button = gr.Button("Transcribe")
yt_output = gr.Textbox(label="Transcription")
file_transcribe_button.click(transcribe_audio, inputs=audio_input, outputs=[file_output, download_file])
yt_transcribe_button.click(transcribe_youtube, inputs=youtube_url, outputs=yt_output)
# Run Gradio app
app.launch(server_name="0.0.0.0", server_port=7880)
|