Spaces:

CORVO-AI
/

TTS

Paused

App Files Files Community

CORVO-AI commited on Mar 10

Commit

202da86

verified ·

1 Parent(s): 7162885

Update app.py

Browse files

Files changed (1) hide show

app.py +32 -37

app.py CHANGED Viewed

@@ -1,6 +1,6 @@
 import os
 import requests
-from flask import Flask, request, jsonify, send_file
 import uuid
 app = Flask(__name__)
@@ -56,8 +56,8 @@ def get_speechify_auth_token(firebase_token):
     else:
         raise Exception(f"❌ Failed to retrieve Speechify token. Status code: {response.status_code}, Response: {response.text}")
-def synthesize_speech(token, text, voice_id, output_file):
-    """Generate speech audio using Speechify API"""
     url = "https://audio.api.speechify.com/v3/synthesis/get"
     headers = {
@@ -73,55 +73,42 @@ def synthesize_speech(token, text, voice_id, output_file):
         "forcedAudioFormat": "mp3"
     }
-    try:
-        response = requests.post(url, headers=headers, json=payload)
         if response.status_code == 200:
-            with open(output_file, "wb") as f:
-                f.write(response.content)
-            print(f"✅ MP3 file saved as {output_file}")
-            return True
         else:
-            print(f"❌ Failed to generate audio. Status code: {response.status_code}")
-            print(response.text)
-            return False
-    except Exception as e:
-        print(f"❌ An error occurred during speech synthesis: {str(e)}")
-        return False
-# Create output directory if it doesn't exist
-os.makedirs("output", exist_ok=True)
 @app.route('/synthesize', methods=['POST'])
 def api_synthesize_speech():
-    """API endpoint to synthesize speech from text"""
     try:
-        # Get parameters from request
         data = request.json
         if not data:
             return jsonify({"error": "No JSON data provided"}), 400
         text = data.get('text')
-        voice_id = data.get('voice_id', 'PVL:4f4a27ef-2b17-424f-904c-30bd1ed60fb8')  # Default voice if not provided
         if not text:
             return jsonify({"error": "No text provided"}), 400
-        # Generate a unique filename
-        filename = f"output/{uuid.uuid4()}.mp3"
-        # Get tokens and synthesize speech
         firebase_token = get_firebase_token()
         speechify_token = get_speechify_auth_token(firebase_token)
-        result = synthesize_speech(speechify_token, text, voice_id, filename)
-        if result:
-            # Return the audio file
-            return send_file(filename, mimetype='audio/mpeg', as_attachment=True)
-        else:
-            return jsonify({"error": "Failed to synthesize speech"}), 500
     except Exception as e:
         return jsonify({"error": str(e)}), 500
@@ -140,8 +127,8 @@ def home():
             </style>
         </head>
         <body>
-            <h1>Speechify Text-to-Speech API</h1>
-            <p>Use this API to convert text to speech using Speechify.</p>
             <h2>Endpoint</h2>
             <code>POST /synthesize</code>
@@ -158,11 +145,19 @@ def home():
             <pre>
 curl -X POST http://localhost:7860/synthesize \\
      -H "Content-Type: application/json" \\
-     -d '{"text": "Hello, this is a test", "voice_id": "PVL:4f4a27ef-2b17-424f-904c-30bd1ed60fb8"}'
             </pre>
             <h2>Response</h2>
-            <p>The API returns an MP3 audio file if successful, or a JSON error message if there's a problem.</p>
         </body>
     </html>
     """

 import os
 import requests
+from flask import Flask, request, jsonify, Response
 import uuid
 app = Flask(__name__)
     else:
         raise Exception(f"❌ Failed to retrieve Speechify token. Status code: {response.status_code}, Response: {response.text}")
+def stream_speech(token, text, voice_id):
+    """Generate speech audio using Speechify API and yield chunks"""
     url = "https://audio.api.speechify.com/v3/synthesis/get"
     headers = {
         "forcedAudioFormat": "mp3"
     }
+    with requests.post(url, headers=headers, json=payload, stream=True) as response:
         if response.status_code == 200:
+            for chunk in response.iter_content(chunk_size=4096):
+                if chunk:
+                    yield chunk
         else:
+            raise Exception(f"Failed to generate audio. Status code: {response.status_code}, Response: {response.text}")
 @app.route('/synthesize', methods=['POST'])
 def api_synthesize_speech():
+    """API endpoint to synthesize speech from text and stream the audio back"""
     try:
         data = request.json
         if not data:
             return jsonify({"error": "No JSON data provided"}), 400
         text = data.get('text')
+        voice_id = data.get('voice_id', 'PVL:4f4a27ef-2b17-424f-904c-30bd1ed60fb8')
         if not text:
             return jsonify({"error": "No text provided"}), 400
+        # Get tokens
         firebase_token = get_firebase_token()
         speechify_token = get_speechify_auth_token(firebase_token)
+        # Stream the audio response
+        return Response(
+            stream_speech(speechify_token, text, voice_id),
+            content_type='audio/mpeg',
+            headers={
+                'Content-Disposition': f'attachment; filename="{uuid.uuid4()}.mp3"',
+                'Transfer-Encoding': 'chunked'
+            }
+        )
     except Exception as e:
         return jsonify({"error": str(e)}), 500
             </style>
         </head>
         <body>
+            <h1>Speechify Text-to-Speech API (Streaming)</h1>
+            <p>Use this API to convert text to speech using Speechify. Audio is streamed back in real-time.</p>
             <h2>Endpoint</h2>
             <code>POST /synthesize</code>
             <pre>
 curl -X POST http://localhost:7860/synthesize \\
      -H "Content-Type: application/json" \\
+     -d '{"text": "Hello, this is a test"}' \\
+     --output speech.mp3
+            </pre>
+            <h2>Example: Stream directly to a media player</h2>
+            <pre>
+curl -X POST http://localhost:7860/synthesize \\
+     -H "Content-Type: application/json" \\
+     -d '{"text": "Hello, this is a streaming test"}' | mpv -
             </pre>
             <h2>Response</h2>
+            <p>The API streams back an MP3 audio file using chunked transfer encoding. No file is saved on the server.</p>
         </body>
     </html>
     """