Spaces:

GamerC0der
/

test-2

Sleeping

App Files Files Community

GamerC0der commited on Jan 8

Commit

f392df9

verified ·

1 Parent(s): 8078753

Update app.py

Browse files

Files changed (1) hide show

app.py +26 -1

app.py CHANGED Viewed

@@ -2,13 +2,19 @@ import http.server
 import socketserver
 import json
 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):
@@ -44,6 +50,25 @@ class Handler(http.server.BaseHTTPRequestHandler):
             self.send_header("Content-type", "application/json")
             self.end_headers()
             self.wfile.write(json.dumps(response).encode())
         else:
             self.send_error(404)

 import socketserver
 import json
 from curl_cffi import requests
+from openai import OpenAI
 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"
+client = OpenAI(
+    base_url="https://integrate.api.nvidia.com/v1",
+    api_key="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"
+)
 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><button onclick="showTab('chat')">Chat (Llama)</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><div id="chatTabContent" style="display:none;"><div id="messages" style="height:300px;overflow-y:scroll;border:1px solid #ccc;padding:10px;margin-bottom:10px;background:#eee;"></div><input type="text" id="chatInput" placeholder="Type message..." style="width:70%;"><button onclick="sendMessage()">Send</button><p>Status: <span id="statusChat">Idle</span></p></div><script>let messages=[];function showTab(tab){document.getElementById('sttTabContent').style.display=tab==='stt'?'block':'none';document.getElementById('ttsTabContent').style.display=tab==='tts'?'block':'none';document.getElementById('chatTabContent').style.display=tab==='chat'?'block':'none';if(tab==='chat'){document.getElementById('chatInput').focus();}}function addMessage(role,content){const div=document.getElementById('messages');const msg=document.createElement('div');msg.innerText=`${role}: ${content}`;div.appendChild(msg);div.scrollTop=div.scrollHeight;}async function sendMessage(){const input=document.getElementById('chatInput');const text=input.value.trim();if(!text)return;addMessage('user',text);input.value='';document.getElementById('statusChat').innerText='Thinking...';messages.push({role:'user',content:text});try{const res=await fetch('/api/chat',{method:'POST',headers:{'Content-Type':'application/json'},body:JSON.stringify({messages:messages})});const data=await res.json();const response=data.response;addMessage('assistant',response);messages.push({role:'assistant',content:response});document.getElementById('statusChat').innerText='Done';}catch(e){document.getElementById('statusChat').innerText='Error';}}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):
             self.send_header("Content-type", "application/json")
             self.end_headers()
             self.wfile.write(json.dumps(response).encode())
+        elif self.path == '/api/chat':
+            content_length = int(self.headers['Content-Length'])
+            body_str = self.rfile.read(content_length).decode('utf-8')
+            req_data = json.loads(body_str)
+            messages = req_data['messages']
+            completion = client.chat.completions.create(
+                model="meta/llama-3.2-1b-instruct",
+                messages=messages,
+                temperature=0.2,
+                top_p=0.7,
+                max_tokens=1024,
+                stream=False
+            )
+            response_text = completion.choices[0].message.content
+            response = {"response": response_text}
+            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)