Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+exhale filter=lfs diff=lfs merge=lfs -text
+ffmpeg filter=lfs diff=lfs merge=lfs -text
+opusenc filter=lfs diff=lfs merge=lfs -text

Dockerfile

@@ -0,0 +1,13 @@
+FROM python:3.12
+
+# RUN useradd -m -u 1000 user
+# USER user
+# ENV PATH="/home/user/.local/bin:$PATH"
+
+WORKDIR /app
+
+RUN pip install flask httpx yt_dlp
+
+COPY . /app
+RUN chmod 777 /app && chmod 777 /app/*
+CMD ["python", "app.py"]

README.md

@@ -1,8 +1,8 @@
 ---
-title: Audiomini
-emoji: 👀
-colorFrom: blue
-colorTo: blue
+title: Audio
+emoji: 🐢
+colorFrom: red
+colorTo: indigo
 sdk: docker
 pinned: false
 ---

app.py

@@ -0,0 +1,172 @@
+from pathlib import Path
+import subprocess
+from re import compile, findall
+from time import sleep
+from typing import Literal
+from urllib.parse import quote_plus
+
+import uuid
+import threading
+
+from flask import Flask, render_template, request, jsonify, send_file
+from httpx import Client
+
+from yt_dlp import YoutubeDL
+
+
+def extract_youtube_id(url: str) -> list[str] | str | None:
+    youtube_regex = r'(https?://)?(www\.)?(youtube|youtu|youtube-nocookie)\.(com|be)/(watch\?v=|embed/|v/|.+\?v=|shorts/)?([^&=%\?]{11})'
+    youtube_url_format = compile(youtube_regex)
+    urls = findall(youtube_url_format, url)
+    if len(urls) > 0:
+        return urls[0][-1]
+    else:
+        return None
+
+
+def download_yt_audio_ytdlp(youtube_id: str) -> Path:
+    ydl_opts = {
+        'format': 'bestaudio/best',
+        'outtmpl': '%(title)s.%(ext)s',
+    }
+    with YoutubeDL(ydl_opts) as ydl:
+        info_dict = ydl.extract_info(f'https://www.youtube.com/watch?v={youtube_id}', download=True)
+        file_name = Path(ydl.prepare_filename(info_dict))
+        return file_name
+
+
+def download_yt_audio_api(youtube_id: str) -> Path:
+    encoded_youtube_url = quote_plus(f'https://www.youtube.com/watch?v={youtube_id}')
+    api_response_url = f'https://ab.cococococ.com/ajax/download.php?copyright=0&format=webm&url={encoded_youtube_url}&api=dfcb6d76f2f6a9894gjkege8a4ab232222'
+
+    with Client() as client:
+        response = client.get(api_response_url)
+        response.raise_for_status()
+        data = response.json()
+        file_name = data['info']['title'].replace(' ', '_')
+        video_id = data.get('id')
+        if not video_id:
+            raise ValueError("ошибка на стороне апи!")
+
+        progress_url = f'https://p.oceansaver.in/ajax/progress.php?id={video_id}'
+        while True:
+            response = client.get(progress_url)
+            response.raise_for_status()
+            data = response.json()
+            download_url = data.get('download_url')
+            if download_url:
+                break
+            sleep(1)
+
+        response = client.get(download_url)
+        response.raise_for_status()
+        path = Path('./')
+        file_name = path / f'{file_name}.webm'
+        with open(file_name, 'wb') as f:
+            f.write(response.content)
+        return file_name
+
+
+def download_yt_audio(youtube_id: str) -> Path | None:
+    try:
+        return download_yt_audio_ytdlp(youtube_id)
+    except Exception as e:
+        print(e)
+        return download_yt_audio_api(youtube_id)
+
+
+def audio_to_wav(input_file: Path) -> Path | None:
+    # if not input_file or not input_file.is_file():
+    #     return None
+    wav_file = f'{input_file.stem}.wav'
+    ffmpeg_command = ["./ffmpeg", "-i", input_file, "-ac", "1", wav_file]
+    subprocess.run(ffmpeg_command, check=True)
+    input_file.unlink(missing_ok=True)
+    return Path(wav_file)
+
+
+def wav_to_opus(wav_file: Path, del_input: bool = True) -> Path:
+    # if not wav_file or not wav_file.is_file():
+    #     raise ValueError('входной файл для кодирования не был получен!')
+    opus_file = f'{wav_file.stem}.opus'
+    opusenc_command = [
+        "./opusenc",
+        "--bitrate", "18",
+        "--vbr",
+        "--speech",
+        "--comp", "10",
+        "--framesize", "60",
+        "--downmix-mono",
+        wav_file, opus_file
+    ]
+    subprocess.run(opusenc_command, check=True)
+    if del_input:
+        wav_file.unlink(missing_ok=True)
+    return Path(opus_file)
+
+
+def wav_to_usac(wav_file: Path, del_input: bool = True) -> Path:
+    # if not wav_file or not wav_file.is_file():
+    #     raise ValueError('входной файл для кодирования не был получен!')
+    exhale_file = f'{wav_file.stem}.m4a'
+    exhale_command = ["./exhale", "0", wav_file, exhale_file]
+    subprocess.run(exhale_command, check=True)
+    if del_input:
+        wav_file.unlink(missing_ok=True)
+    return Path(exhale_file)
+
+
+def download_and_encode(youtube_url: str, codec: Literal['opus', 'usac', 'both'] = 'both') -> list[Path]:
+    temp_wav = audio_to_wav(download_yt_audio(extract_youtube_id(youtube_url)))
+    if codec == 'opus':
+        return [wav_to_opus(temp_wav)]
+    elif codec == 'usac':
+        return [wav_to_usac(temp_wav)]
+    else:
+        return [wav_to_opus(temp_wav, del_input=False), wav_to_usac(temp_wav)]
+
+
+app = Flask(__name__)
+
+
+tasks = {}
+
+@app.route('/')
+def index():
+    return render_template('index.html')
+
+@app.route('/download', methods=['POST'])
+def download():
+    youtube_url = request.form['youtube_url']
+    codec = request.form['codec']
+    task_id = str(uuid.uuid4())
+
+    def process_task():
+        tasks[task_id] = {'status': 'скачивание с ютуба...'}
+        audio_file = download_yt_audio(extract_youtube_id(youtube_url))
+        tasks[task_id]['status'] = 'предварительная конвертация...'
+        wav_file = audio_to_wav(audio_file)
+        tasks[task_id]['status'] = 'сжатие...'
+        encoded_files = download_and_encode(youtube_url, codec)
+        tasks[task_id]['status'] = 'готово'
+        tasks[task_id]['files'] = [str(f) for f in encoded_files]
+
+    threading.Thread(target=process_task).start()
+    return jsonify({'task_id': task_id})
+
+@app.route('/download_status/<task_id>')
+def download_status(task_id):
+    task = tasks.get(task_id)
+    if task:
+        return jsonify(task)
+    else:
+        return jsonify({'error': 'Task not found'}), 404
+
+
+@app.route('/play/<path:filename>')
+def play(filename):
+    return send_file(filename, as_attachment=True)
+
+
+if __name__ == '__main__':
+    app.run(host='0.0.0.0', port=7860)

exhale

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd563aed06aa623d35364f890b7ba194fafcd5c5e29fd36b2108b1d1f256410e
+size 1097864

ffmpeg

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5267ed734dc7b2f5955f93924a3d89f0fb43f8fbcc8f0da183bc6a2edde3ec5
+size 125709160

opusenc

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d52aef7d08218ae89c2cbc775f2d32f6f1276704c907c75a47f774ae120a8086
+size 2892712

templates/index.html

@@ -0,0 +1,171 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <title>Audio Downloader</title>
+    <style>
+      body {
+          background-color: #121212;
+          color: #e0e0e0;
+          font-family: 'Segoe UI', Tahoma, Verdana, sans-serif;
+          margin: 0;
+          padding: 20px;
+      }
+      
+      h1 {
+          color: #59a459;
+          font-size: 2em;
+          margin-bottom: 20px;
+          display: none;
+      }
+      
+      #downloadForm {
+          display: flex;
+          flex-direction: column;
+          flex-wrap: wrap;
+          align-content: center;
+          justify-content: center;
+          align-items: center;
+      }
+      
+      form {
+          background-color: #1f1f1f;
+          border-radius: 8px;
+          padding: 20px;
+          box-shadow: 0 4px 8px rgba(0, 0, 0, 0.2);
+      }
+      
+      div.codec {
+          display: none;
+      }
+      
+      label {
+          font-size: 1.1em;
+          margin-bottom: 8px;
+      }
+      
+      label[for="youtube_url"] {
+          display: none
+      }
+      
+      input[type="text"] {
+          width: 40%;
+          padding: 10px;
+          margin-bottom: 20px;
+          border: 1px solid #3c3c3c;
+          border-radius: 4px;
+          background-color: #2c2c2c;
+          color: #e0e0e0;
+      }
+      
+      input[type="radio"] {
+          margin-right: 8px;
+      }
+      
+      input[type="radio"] + label {
+          margin-right: 20px;
+          font-size: 1em;
+      }
+      
+      button {
+          margin-top: 20px;
+          width: 40%;
+          background-color: #86e598;
+          color: #121212;
+          border: none;
+          padding: 10px 20px;
+          border-radius: 4px;
+          font-size: 1.1em;
+          cursor: pointer;
+          transition: background-color 0.3s;
+      }
+      
+      button:hover {
+          background-color: #58cc4d;
+      }
+      
+      #status {
+          margin-top: 20px;
+          font-size: 1.1em;
+      }
+      
+      a {
+          color: #03da84;
+          text-decoration: none;
+          transition: color 0.3s;
+      }
+      
+      a:hover {
+          color: #0ef758;
+      }
+      
+      audio {
+          margin-top: 10px;
+          width: 100%;
+      }
+    </style>
+    <script>
+        function pollTaskStatus(task_id) {
+            fetch(`/download_status/${task_id}`)
+                .then(response => response.json())
+                .then(data => {
+                    document.getElementById('status').innerText = data.status;
+                    if (data.status === 'готово') {
+                        data.files.forEach(file => {
+                            const p = document.createElement('p');
+                            const a = document.createElement('a');
+                            a.href = `/play/${file}`;
+                            a.innerText = file;
+                            p.innerText = 'File: ';
+                            p.appendChild(a);
+                            document.body.appendChild(p);
+                            const audio = document.createElement('audio');
+                            audio.controls = true;
+                            const source = document.createElement('source');
+                            source.src = `/play/${file}`;
+                            source.type = 'audio/mpeg';
+                            audio.appendChild(source);
+                            document.body.appendChild(audio);
+                        });
+                    } else {
+                        setTimeout(() => pollTaskStatus(task_id), 1000);
+                    }
+                });
+        }
+
+        function startDownload() {
+            const form = document.getElementById('downloadForm');
+            const formData = new FormData(form);
+            fetch('/download', {
+                method: 'POST',
+                body: formData
+            })
+            .then(response => response.json())
+            .then(data => {
+                document.getElementById('status').innerText = 'Starting download...';
+                pollTaskStatus(data.task_id);
+            });
+        }
+    </script>
+</head>
+<body>
+    <h1>получить минимальную звуковую дорожку с YouTube</h1>
+    <form id="downloadForm">
+        <label for="youtube_url">ссылка на видео:</label>
+        <input type="text" id="youtube_url" name="youtube_url" placeholder="ссылка на видео" required>
+        <br>
+        <div id="status"></div>
+        <div class="codec">
+        <label for="codec">кодек:</label>
+        <input type="radio" id="opus" name="codec" value="opus" checked>
+        <label for="opus">Opus</label>
+        <input type="radio" id="usac" name="codec" value="usac">
+        <label for="usac">USAC</label>
+        <input type="radio" id="both" name="codec" value="both">
+        <label for="both">Both</label>
+        </div>
+        <br>
+        <button type="button" onclick="startDownload()">Download</button>
+    </form>
+</body>
+</html>