Spaces:

awacke1
/

RescuerOfStolenBikes

Running

App Files Files Community

awacke1 commited on 24 days ago

Commit

24d6aa2

•

1 Parent(s): 460cd22

Update app.py

Browse files

Files changed (1) hide show

app.py +174 -108

app.py CHANGED Viewed

@@ -519,6 +519,169 @@ def display_file_manager():
                 os.remove(file)
                 st.rerun()
 # Helper Functions
 def generate_filename(prompt, file_type):
     central = pytz.timezone('US/Central')
@@ -652,6 +815,7 @@ def set_transcript(text):
 def main():
     st.sidebar.markdown("### 🚲BikeAI🏆 Claude and GPT Multi-Agent Research AI")
     tab_main = st.radio("Choose Action:",
                         ["🎤 Voice Input", "📸 Media Gallery", "🔍 Search ArXiv", "📝 File Editor"],
                         horizontal=True)
@@ -659,119 +823,21 @@ def main():
     if tab_main == "🎤 Voice Input":
         st.subheader("Voice Recognition")
-        # Initialize session state for transcript
         if 'voice_transcript' not in st.session_state:
             st.session_state.voice_transcript = ""
-        # Add a container to display the transcript
-        transcript_container = st.empty()
-        # Modify the JavaScript to use Streamlit's component communication
-        # Create a simpler HTML component that just updates a value we can read
-        speech_recognition_html = """
-        <!DOCTYPE html>
-        <html>
-        <head>
-            <title>Speech Recognition</title>
-        </head>
-        <body>
-            <div>
-                <button id="startButton">Start</button>
-                <button id="stopButton" disabled>Stop</button>
-                <div id="status">Click Start to begin</div>
-                <div id="output"></div>
-            </div>
-            <script>
-                let recognition;
-                let transcript = '';
-                function initSpeechRecognition() {
-                    if ('webkitSpeechRecognition' in window) {
-                        recognition = new webkitSpeechRecognition();
-                        recognition.continuous = true;
-                        recognition.interimResults = true;
-                        recognition.onresult = (event) => {
-                            let interimTranscript = '';
-                            let finalTranscript = '';
-                            for (let i = event.resultIndex; i < event.results.length; i++) {
-                                const result = event.results[i][0].transcript;
-                                if (event.results[i].isFinal) {
-                                    finalTranscript += result;
-                                } else {
-                                    interimTranscript += result;
-                                }
-                            }
-                            if (finalTranscript) {
-                                transcript += finalTranscript + ' ';
-                                document.getElementById('output').innerText = transcript;
-                                // Set this as the return value for Streamlit
-                                document.getElementById('streamlit-data').value = transcript;
-                            }
-                        };
-                        recognition.onend = () => {
-                            if (!document.getElementById('stopButton').disabled) {
-                                recognition.start();
-                            }
-                        };
-                    }
-                }
-                document.getElementById('startButton').onclick = () => {
-                    recognition.start();
-                    document.getElementById('startButton').disabled = true;
-                    document.getElementById('stopButton').disabled = false;
-                    document.getElementById('status').innerText = 'Listening...';
-                };
-                document.getElementById('stopButton').onclick = () => {
-                    recognition.stop();
-                    document.getElementById('startButton').disabled = false;
-                    document.getElementById('stopButton').disabled = true;
-                    document.getElementById('status').innerText = 'Stopped';
-                };
-                // Initialize on load
-                window.onload = () => {
-                    initSpeechRecognition();
-                };
-            </script>
-            <!-- Hidden input for Streamlit to read -->
-            <input type="hidden" id="streamlit-data" value="">
-        </body>
-        </html>
-        """
-        # Display the component
-        components.html(speech_recognition_html, height=400)
-        # Add a placeholder for the transcript
-        transcript_placeholder = st.empty()
-        # Add a refresh button
-        if st.button("Update Transcript"):
-            st.rerun()
-        # Display the current transcript from session state
-        if 'voice_transcript' in st.session_state:
-            transcript_placeholder.text_area(
-                "Transcript:",
-                value=st.session_state.voice_transcript,
-                height=150
-            )
         # Model Selection
         model_choice = st.sidebar.radio(
             "Choose AI Model:",

                 os.remove(file)
                 st.rerun()
+# Speech Recognition HTML Component
+speech_recognition_html = """
+<!DOCTYPE html>
+<html>
+<head>
+    <title>Continuous Speech Demo</title>
+    <style>
+        body {
+            font-family: sans-serif;
+            padding: 20px;
+            max-width: 800px;
+            margin: 0 auto;
+        }
+        button {
+            padding: 10px 20px;
+            margin: 10px 5px;
+            font-size: 16px;
+        }
+        #status {
+            margin: 10px 0;
+            padding: 10px;
+            background: #e8f5e9;
+            border-radius: 4px;
+        }
+        #output {
+            white-space: pre-wrap;
+            padding: 15px;
+            background: #f5f5f5;
+            border-radius: 4px;
+            margin: 10px 0;
+            min-height: 100px;
+            max-height: 400px;
+            overflow-y: auto;
+        }
+        .controls {
+            margin: 10px 0;
+        }
+    </style>
+</head>
+<body>
+    <div class="controls">
+        <button id="start">Start Listening</button>
+        <button id="stop" disabled>Stop Listening</button>
+        <button id="clear">Clear Text</button>
+    </div>
+    <div id="status">Ready</div>
+    <div id="output"></div>
+    <!-- Add the hidden input here -->
+    <input type="hidden" id="streamlit-data" value="">
+    <script>
+        if (!('webkitSpeechRecognition' in window)) {
+            alert('Speech recognition not supported');
+        } else {
+            const recognition = new webkitSpeechRecognition();
+            const startButton = document.getElementById('start');
+            const stopButton = document.getElementById('stop');
+            const clearButton = document.getElementById('clear');
+            const status = document.getElementById('status');
+            const output = document.getElementById('output');
+            let fullTranscript = '';
+            let lastUpdateTime = Date.now();
+            // Configure recognition
+            recognition.continuous = true;
+            recognition.interimResults = true;
+            // Function to start recognition
+            const startRecognition = () => {
+                try {
+                    recognition.start();
+                    status.textContent = 'Listening...';
+                    startButton.disabled = true;
+                    stopButton.disabled = false;
+                } catch (e) {
+                    console.error(e);
+                    status.textContent = 'Error: ' + e.message;
+                }
+            };
+            // Auto-start on load
+            window.addEventListener('load', () => {
+                setTimeout(startRecognition, 1000);
+            });
+            startButton.onclick = startRecognition;
+            stopButton.onclick = () => {
+                recognition.stop();
+                status.textContent = 'Stopped';
+                startButton.disabled = false;
+                stopButton.disabled = true;
+            };
+            clearButton.onclick = () => {
+                fullTranscript = '';
+                output.textContent = '';
+                window.parent.postMessage({
+                    type: 'clear_transcript',
+                }, '*');
+            };
+            recognition.onresult = (event) => {
+                let interimTranscript = '';
+                let finalTranscript = '';
+                for (let i = event.resultIndex; i < event.results.length; i++) {
+                    const transcript = event.results[i][0].transcript;
+                    if (event.results[i].isFinal) {
+                        finalTranscript += transcript + '\\n';
+                    } else {
+                        interimTranscript += transcript;
+                    }
+                }
+                if (finalTranscript || (Date.now() - lastUpdateTime > 5000)) {
+                    if (finalTranscript) {
+                        fullTranscript += finalTranscript;
+                        // Update the hidden input value
+                        document.getElementById('streamlit-data').value = fullTranscript;
+                    }
+                    lastUpdateTime = Date.now();
+                }
+                output.textContent = fullTranscript + (interimTranscript ? '... ' + interimTranscript : '');
+                output.scrollTop = output.scrollHeight;
+                document.getElementById('streamlit-data').value = fullTranscript;
+            };
+            recognition.onend = () => {
+                if (!stopButton.disabled) {
+                    try {
+                        recognition.start();
+                        console.log('Restarted recognition');
+                    } catch (e) {
+                        console.error('Failed to restart recognition:', e);
+                        status.textContent = 'Error restarting: ' + e.message;
+                        startButton.disabled = false;
+                        stopButton.disabled = true;
+                    }
+                }
+            };
+            recognition.onerror = (event) => {
+                console.error('Recognition error:', event.error);
+                status.textContent = 'Error: ' + event.error;
+                if (event.error === 'not-allowed' || event.error === 'service-not-allowed') {
+                    startButton.disabled = false;
+                    stopButton.disabled = true;
+                }
+            };
+        }
+    </script>
+</body>
+</html>
+"""
 # Helper Functions
 def generate_filename(prompt, file_type):
     central = pytz.timezone('US/Central')
 def main():
     st.sidebar.markdown("### 🚲BikeAI🏆 Claude and GPT Multi-Agent Research AI")
+    # Main navigation
     tab_main = st.radio("Choose Action:",
                         ["🎤 Voice Input", "📸 Media Gallery", "🔍 Search ArXiv", "📝 File Editor"],
                         horizontal=True)
     if tab_main == "🎤 Voice Input":
         st.subheader("Voice Recognition")
+        # Initialize session state for the transcript
         if 'voice_transcript' not in st.session_state:
             st.session_state.voice_transcript = ""
+        # Display speech recognition component and capture returned value
+        transcript = st.components.v1.html(speech_recognition_html, height=400)
+        # Update session state if there's new data
+        if transcript is not None and transcript != "":
+            st.session_state.voice_transcript = transcript
+        # Display the transcript in a Streamlit text area
+        # st.markdown("### Processed Voice Input:")
+        # st.text_area("Voice Transcript", st.session_state.voice_transcript, height=100)
         # Model Selection
         model_choice = st.sidebar.radio(
             "Choose AI Model:",