Update audio_generator.py

audio_generator.py

@@ -1,53 +1,54 @@
 import os
 import edge_tts
 import asyncio
-from elevenlabs import generate, save, voices
-from elevenlabs.api.error import APIError
+from typing import Optional
+
+# Available free voices (natural-sounding options)
+VOICES = {
+    "female": {
+        "en-US": "en-US-AriaNeural",  # Very natural
+        "en-GB": "en-GB-LibbyNeural",  # British English
+        "es-ES": "es-ES-ElviraNeural"  # Spanish
+    },
+    "male": {
+        "en-US": "en-US-GuyNeural",
+        "en-GB": "en-GB-RyanNeural",
+        "es-ES": "es-ES-AlvaroNeural"
+    }
+}
+
+async def generate_speech(
+    text: str,
+    output_file: str = "output.mp3",
+    voice: Optional[str] = None,
+    rate: str = "+0%",  # Speed adjustment
+    volume: str = "+0%"  # Volume adjustment
+) -> str:
+    """
+    Generate speech using free EdgeTTS
+    Args:
+        text: Input text (max 3000 chars)
+        output_file: Output path
+        voice: Voice code (e.g. 'en-US-AriaNeural')
+        rate: Speaking rate adjustment (-50% to +100%)
+        volume: Volume adjustment (-50% to +50%)
+    """
+    if not voice:
+        voice = VOICES["female"]["en-US"]  # Default to natural female voice
 
-async def generate_edge_audio(text, filename="output_audio.mp3"):
-    """Free Microsoft Edge TTS implementation"""
     try:
-        communicate = edge_tts.Communicate(text, "en-US-AriaNeural")
-        await communicate.save(filename)
-        return filename
+        communicate = edge_tts.Communicate(
+            text=text,
+            voice=voice,
+            rate=rate,
+            volume=volume
+        )
+        await communicate.save(output_file)
+        return output_file
     except Exception as e:
         raise RuntimeError(f"EdgeTTS error: {str(e)}")
 
-def generate_audio(text, filename="output_audio.mp3"):
-    """Hybrid audio generator with fallback logic"""
-    # First try EdgeTTS (free)
-    try:
-        return asyncio.run(generate_edge_audio(text, filename))
-    except Exception as edge_error:
-        # If Edge fails, try ElevenLabs if configured
-        elevenlabs_key = os.getenv("ELEVENLABS_API_KEY")
-        if not elevenlabs_key:
-            raise RuntimeError(
-                "EdgeTTS failed and no ElevenLabs API key configured. "
-                f"Original error: {str(edge_error)}"
-            )
-        
-        try:
-            # Verify ElevenLabs key
-            available_voices = voices()
-            if not available_voices:
-                raise RuntimeError("No ElevenLabs voices available")
-            
-            audio = generate(
-                text=text,
-                voice=available_voices[0],
-                model="eleven_monolingual_v1",
-                api_key=elevenlabs_key
-            )
-            save(audio, filename)
-            return filename
-        except APIError as e:
-            if "requires a valid API key" in str(e):
-                raise RuntimeError("Invalid ElevenLabs API key")
-            raise RuntimeError(f"ElevenLabs error: {str(e)}")
-        except Exception as e:
-            raise RuntimeError(
-                f"Both EdgeTTS and ElevenLabs failed. "
-                f"Edge error: {str(edge_error)}. "
-                f"ElevenLabs error: {str(e)}"
-            )
+def generate_audio(text: str, filename: str = "output.mp3") -> str:
+    """Synchronous wrapper for async TTS"""
+    os.makedirs(os.path.dirname(filename) or ".", exist_ok=True)
+    return asyncio.run(generate_speech(text, filename))