Spaces:

GamerC0der
/

test-2

Sleeping

App Files Files Community

GamerC0der commited on 28 days ago

Commit

86eaf41

verified ·

1 Parent(s): 001b27b

Update app.py

Browse files

Files changed (1) hide show

app.py +30 -34

app.py CHANGED Viewed

@@ -4,52 +4,48 @@ import json
 from curl_cffi import requests
 PORT = 7860
-URL = "https://multi-modal.ai.cloudflare.com/api/inference?model=@cf/deepgram/nova-3&field=audio"
 HTML = """
-<!DOCTYPE html>
-<html>
-<body>
-    <h1>Nova-3 Playground</h1>
-    <input type="file" id="f" accept="audio/mpeg">
-    <button onclick="u()">Transcribe</button>
-    <p>Status: <span id="s">Idle</span></p>
-    <pre id="r" style="background:#eee;padding:10px"></pre>
-    <script>
-        async function u() {
-            const f = document.getElementById('f').files[0];
-            if(!f) return;
-            document.getElementById('s').innerText = "Processing...";
-            try {
-                const res = await fetch('/api', { method: 'POST', body: f });
-                const data = await res.json();
-                document.getElementById('r').innerText = JSON.stringify(data, null, 2);
-                document.getElementById('s').innerText = "Done";
-            } catch(e) {
-                document.getElementById('s').innerText = "Error";
-            }
-        }
-    </script>
-</body>
-</html>
 """
 class Handler(http.server.BaseHTTPRequestHandler):
     def do_GET(self):
-        self.send_response(200)
-        self.send_header("Content-type", "text/html")
-        self.end_headers()
-        self.wfile.write(HTML.encode())
     def do_POST(self):
-        if self.path == '/api':
-            length = int(self.headers['Content-Length'])
-            body = self.rfile.read(length)
-            r = requests.post(URL, data=body, impersonate="chrome")
             self.send_response(200)
             self.send_header("Content-type", "application/json")
             self.end_headers()
             self.wfile.write(r.content)
 with socketserver.TCPServer(("", PORT), Handler) as d:
     print(f"Server: {PORT}")

 from curl_cffi import requests
 PORT = 7860
+STT_URL = "https://multi-modal.ai.cloudflare.com/api/inference?model=@cf/deepgram/nova-3&field=audio"
+TTS_URL = "https://multi-modal.ai.cloudflare.com/api/inference"
 HTML = """
+<!DOCTYPE html><html><body><h1>Multi-Modal Playground</h1><div id="tabs"><button onclick="showTab('stt')">STT (Nova-3)</button><button onclick="showTab('tts')">TTS (AURA-1)</button></div><div id="sttTabContent" style="display:block;"><p>Upload audio:</p><input type="file" id="audioFile" accept="audio/*"><button onclick="transcribe()">Transcribe</button><p>Status: <span id="status">Idle</span></p><pre id="result" style="background:#eee;padding:10px"></pre></div><div id="ttsTabContent" style="display:none;"><p>Enter text:</p><input type="text" id="textInput" placeholder="Enter text to speak" style="width:100%;"><button onclick="generateAudio()">Generate Audio</button><p>Status: <span id="statusTTS">Idle</span></p><audio id="audioPlayer" controls style="width:100%;"></audio></div><script>function showTab(tab){document.getElementById('sttTabContent').style.display=tab==='stt'?'block':'none';document.getElementById('ttsTabContent').style.display=tab==='tts'?'block':'none';}async function transcribe(){const file=document.getElementById('audioFile').files[0];if(!file)return;document.getElementById('status').innerText='Processing...';try{const res=await fetch('/api/stt',{method:'POST',body:file});const data=await res.json();document.getElementById('result').innerText=JSON.stringify(data,null,2);document.getElementById('status').innerText='Done';}catch(e){document.getElementById('status').innerText='Error';}}async function generateAudio(){const text=document.getElementById('textInput').value;if(!text)return;document.getElementById('statusTTS').innerText='Generating...';try{const res=await fetch('/api/tts',{method:'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify({text:text})});const data=await res.json();document.getElementById('audioPlayer').src='data:audio/webm;base64,'+data.audio;document.getElementById('statusTTS').innerText='Done';}catch(e){document.getElementById('statusTTS').innerText='Error';}}</script></body></html>
 """
 class Handler(http.server.BaseHTTPRequestHandler):
     def do_GET(self):
+        if self.path == '/':
+            self.send_response(200)
+            self.send_header("Content-type", "text/html")
+            self.end_headers()
+            self.wfile.write(HTML.encode())
+        else:
+            self.send_error(404)
     def do_POST(self):
+        if self.path == '/api/stt':
+            content_length = int(self.headers['Content-Length'])
+            body = self.rfile.read(content_length)
+            r = requests.post(STT_URL, data=body, impersonate="chrome")
             self.send_response(200)
             self.send_header("Content-type", "application/json")
             self.end_headers()
             self.wfile.write(r.content)
+        elif self.path == '/api/tts':
+            content_length = int(self.headers['Content-Length'])
+            body_str = self.rfile.read(content_length).decode('utf-8')
+            req_data = json.loads(body_str)
+            text = req_data['text']
+            tts_payload = {"model": "@cf/myshell-ai/melotts", "params": {"prompt": text}}
+            r = requests.post(TTS_URL, json=tts_payload, impersonate="chrome")
+            resp_data = r.json()
+            audio_b64 = resp_data["response"]["audio"]
+            response = {"audio": audio_b64}
+            self.send_response(200)
+            self.send_header("Content-type", "application/json")
+            self.end_headers()
+            self.wfile.write(json.dumps(response).encode())
+        else:
+            self.send_error(404)
 with socketserver.TCPServer(("", PORT), Handler) as d:
     print(f"Server: {PORT}")