Rename tts_script.py to app.py

app.py

@@ -0,0 +1,89 @@
+import time
+import requests
+from io import BytesIO
+from pathlib import Path
+from typing import List
+import re
+from flask import Flask, request, render_template, send_file
+
+app = Flask(__name__)
+
+class SentenceTokenizer:
+    """Advanced sentence tokenizer with support for complex cases."""
+    
+    def __init__(self):
+        self.SENTENCE_END = re.compile(
+            r'(?<=[.!?])\s+(?=[A-Z])|(?<=[。！？])\s+',
+            re.VERBOSE
+        )
+
+    def tokenize(self, text: str) -> List[str]:
+        if not text or not text.strip():
+            return []
+        # Simple sentence splitting
+        sentences = self.SENTENCE_END.split(text.strip())
+        return [s.strip() for s in sentences if s.strip()]
+
+def split_sentences(text: str) -> List[str]:
+    tokenizer = SentenceTokenizer()
+    return tokenizer.tokenize(text)
+
+class ElevenlabsTTS:
+    """Text-to-speech provider using Elevenlabs API."""
+    
+    def __init__(self):
+        self.session = requests.Session()
+        self.session.headers.update({"User-Agent": "Mozilla/5.0"})
+        self.cache_dir = Path("./audio_cache")
+        self.all_voices = {
+            "Brian": "nPczCjzI2devNBz1zQrb",
+            "Alice": "Xb7hH8MSUJpSbSDYk0k2",
+            # Add other voices as needed
+        }
+        self.params = {'allow_unauthenticated': '1'}
+
+    def tts(self, text: str, voice: str = "Brian") -> str:
+        if voice not in self.all_voices:
+            raise ValueError(f"Voice '{voice}' not available")
+
+        filename = self.cache_dir / f"{int(time.time())}.mp3"
+        sentences = split_sentences(text)
+
+        audio_chunks = {}
+        for i, sentence in enumerate(sentences, 1):
+            json_data = {'text': sentence, 'model_id': 'eleven_multilingual_v2'}
+            response = self.session.post(
+                f'https://api.elevenlabs.io/v1/text-to-speech/{self.all_voices[voice]}',
+                params=self.params,
+                json=json_data,
+                timeout=20
+            )
+            response.raise_for_status()
+            audio_chunks[i] = response.content
+
+        self.cache_dir.mkdir(parents=True, exist_ok=True)
+        combined_audio = BytesIO()
+        for i in sorted(audio_chunks.keys()):
+            combined_audio.write(audio_chunks[i])
+
+        with open(filename, 'wb') as f:
+            f.write(combined_audio.getvalue())
+        return filename.as_posix()
+
+# Web Interface
+tts_provider = ElevenlabsTTS()
+
+@app.route('/', methods=['GET', 'POST'])
+def index():
+    if request.method == 'POST':
+        text = request.form.get('text')
+        voice = request.form.get('voice', 'Brian')
+        try:
+            audio_file = tts_provider.tts(text, voice)
+            return send_file(audio_file, mimetype='audio/mpeg', as_attachment=True)
+        except Exception as e:
+            return render_template('index.html', error=str(e), voices=tts_provider.all_voices.keys())
+    return render_template('index.html', voices=tts_provider.all_voices.keys())
+
+if __name__ == "__main__":
+    app.run(host='0.0.0.0', port=5000)

tts_script.py

@@ -1,81 +0,0 @@
-import time
-import requests
-import pathlib
-from io import BytesIO
-from flask import Flask, request, render_template, send_file
-from pydub import AudioSegment
-
-# Flask App Setup
-app = Flask(__name__)
-
-# ElevenLabs API Configuration
-ELEVENLABS_API_URL = "https://api.elevenlabs.io/v1/text-to-speech"
-HEADERS = {"User-Agent": "TTSApp"}
-CACHE_DIR = pathlib.Path("/tmp/audio_cache")
-CACHE_DIR.mkdir(parents=True, exist_ok=True)
-
-# Available Voices
-ALL_VOICES = {
-    "Brian": "nPczCjzI2devNBz1zQrb",
-    "Alice": "Xb7hH8MSUJpSbSDYk0k2",
-    "Will": "bIHbv24MWmeRgasZH58o",
-}
-
-# Split text into sentences (Basic)
-def split_sentences(text):
-    return text.split(". ")
-
-# Generate TTS
-def generate_audio(text, voice):
-    if voice not in ALL_VOICES:
-        return {"error": f"Invalid voice '{voice}'"}
-
-    filename = CACHE_DIR / f"{int(time.time())}.mp3"
-    sentences = split_sentences(text)
-
-    audio_chunks = []
-    for sentence in sentences:
-        response = requests.post(
-            f"{ELEVENLABS_API_URL}/{ALL_VOICES[voice]}",
-            headers=HEADERS,
-            json={"text": sentence, "model_id": "eleven_multilingual_v2"},
-            timeout=20
-        )
-        if response.ok:
-            audio_chunks.append(BytesIO(response.content))
-
-    # Combine all audio parts
-    combined_audio = AudioSegment.empty()
-    for chunk in audio_chunks:
-        chunk_audio = AudioSegment.from_file(chunk, format="mp3")
-        combined_audio += chunk_audio
-
-    combined_audio.export(filename, format="mp3")
-    return filename.as_posix()
-
-# Flask Routes
-@app.route("/", methods=["GET", "POST"])
-def home():
-    if request.method == "POST":
-        text = request.form["text"]
-        voice = request.form.get("voice", "Brian")
-        audio_file = generate_audio(text, voice)
-        return send_file(audio_file, as_attachment=True)
-
-    return """
-    <h1>Text-to-Speech Generator</h1>
-    <form method="post">
-        <label>Text:</label><br>
-        <textarea name="text" rows="4" cols="50" required></textarea><br>
-        <label>Voice:</label>
-        <select name="voice">
-            <option value="Brian">Brian</option>
-            <option value="Alice">Alice</option>
-            <option value="Will">Will</option>
-        </select><br><br>
-        <button type="submit">Generate & Download</button>
-    </form>
-    """
-
-if __name__ == "__main__":
-    app.run(debug=True, host="0.0.0.0", port=5000)