У меня не так много опыта работы с JavaScript.
Мне нужно выполнить задачу, в которой я должен выполнить как STT, так и TTS. Я использую речь в библиотеке JavaScript Syenthesis uTERENE, эта библиотека хорошо работает при тестировании с устройством Android, но преобразование речи в текстовое преобразование не удалось во время его использования с устройством iOS. < /P>
Voice Command
.chat-container {
max-width: 400px;
margin: 20px auto;
padding: 10px;
border: 1px solid #ccc;
border-radius: 5px;
box-shadow: 0 0 10px rgba(0, 0, 0, 0.1);
font-family: Arial, sans-serif;
}
.user-message {
background-color: #f0f0f0;
border-radius: 5px;
padding: 5px 10px;
margin: 5px 0;
text-align: right;
}
.bot-message {
background-color: #d3e9ff;
border-radius: 5px;
padding: 5px 10px;
margin: 5px 0;
}
#languageSelector {
width: 100%;
margin-top: 10px;
padding: 5px;
border-radius: 5px;
border: 1px solid #ccc;
}
#status {
color: grey;
font-weight: 600;
margin-top: 10px;
text-align: center;
}
#permissionModal {
position: fixed;
top: 0;
left: 0;
width: 100%;
height: 100%;
background: rgba(0, 0, 0, 0.5);
display: flex;
justify-content: center;
align-items: center;
z-index: 1000;
}
#permissionModal div {
background: white;
padding: 20px;
border-radius: 5px;
text-align: center;
box-shadow: 0 0 10px rgba(0, 0, 0, 0.3);
}
#permissionModal button {
margin: 10px;
padding: 10px 20px;
border: none;
border-radius: 5px;
background: #007bff;
color: white;
cursor: pointer;
}
#permissionModal button:hover {
background: #0056b3;
}
#testSpeakerButton {
display: block;
margin: 10px auto;
padding: 10px 20px;
border: none;
border-radius: 5px;
background: #28a745;
color: white;
cursor: pointer;
font-family: Arial, sans-serif;
font-weight: 600;
}
#testSpeakerButton:hover {
background: #218838;
}
This site requires microphone and speaker permissions to enable voice input and output.
Grant Permissions
Speaker test
English (US)
Hindi (India)
Spanish (Spain)
French (France)
German (Germany)
Arabic (Saudi Arabia)
Tap to Speak
// Browser detection
function detectBrowser() {
const ua = navigator.userAgent.toLowerCase();
if (ua.includes('safari') && !ua.includes('chrome')) return 'Safari';
if (ua.includes('firefox')) return 'Firefox';
if (ua.includes('edg')) return 'Edge';
if (ua.includes('chrome')) return 'Chrome';
return 'Unknown';
}
const browser = detectBrowser();
const statusBar = document.getElementById('status');
const permissionModal = document.getElementById('permissionModal');
const grantPermissionsButton = document.getElementById('grantPermissions');
const testSpeakerButton = document.getElementById('testSpeakerButton');
// Language mapping
const speechLangMap = {
'English (US)': 'en-US',
'Hindi (India)': 'hi-IN',
'Spanish (Spain)': 'es-ES',
'French (France)': 'fr-FR',
'German (Germany)': 'de-DE',
'Arabic (Saudi Arabia)': 'ar-SA'
};
// Initialize speech synthesis
const synth = window.speechSynthesis || null;
let voices = [];
// Load voices asynchronously
function loadVoices() {
return new Promise((resolve) => {
if (!synth) {
statusBar.textContent = 'Text-to-speech not supported in this browser.';
resolve([]);
return;
}
voices = synth.getVoices();
if (voices.length > 0) {
resolve(voices);
} else {
synth.addEventListener('voiceschanged', () => {
voices = synth.getVoices();
resolve(voices);
}, { once: true });
}
});
}
// Speak text with fallback
async function speakResponse(text, language) {
if (!synth) {
statusBar.textContent = 'Text-to-speech is unavailable. Displaying text only.';
showBotMessage(text);
return;
}
const langCode = speechLangMap[language] || 'en-US';
await loadVoices();
const utterance = new SpeechSynthesisUtterance(text);
let selectedVoice = voices.find(voice => voice.lang === langCode);
if (!selectedVoice) {
console.warn(`No voice for ${langCode}. Falling back to English.`);
selectedVoice = voices.find(voice => voice.lang.startsWith('en')) || voices[0];
statusBar.textContent = `Voice for ${language} unavailable. Using English voice.`;
}
if (selectedVoice) {
utterance.voice = selectedVoice;
utterance.lang = selectedVoice.lang;
} else {
statusBar.textContent = 'No voices available for text-to-speech.';
showBotMessage(text);
return;
}
utterance.volume = 1.0;
utterance.rate = 1.0;
utterance.pitch = 1.0;
utterance.onerror = (event) => {
console.error('TTS error:', event.error);
statusBar.textContent = 'Error in text-to-speech. Displaying text only.';
showBotMessage(text);
};
utterance.onend = () => {
console.log('TTS finished.');
statusBar.textContent = '';
};
if (synth.speaking || synth.paused) {
synth.cancel();
}
try {
synth.speak(utterance);
} catch (error) {
console.error('TTS failed:', error);
statusBar.textContent = 'Failed to play speech. Displaying text only.';
showBotMessage(text);
}
document.getElementById('speech').addEventListener('click', () => {
if (synth.speaking) synth.cancel();
}, { once: true });
}
// Initialize TTS and test speaker
async function testSpeaker() {
if (!synth) {
statusBar.textContent = 'Text-to-speech not supported in this browser.';
return false;
}
try {
await loadVoices();
// Silent utterance to unlock audio in Safari
const silentUtterance = new SpeechSynthesisUtterance('');
silentUtterance.volume = 0;
silentUtterance.onend = () => synth.cancel();
silentUtterance.onerror = (event) => {
console.error('Silent TTS error:', event.error);
synth.cancel();
};
synth.speak(silentUtterance);
await new Promise(resolve => setTimeout(resolve, 100));
synth.cancel();
// Test utterance
const selectedLang = document.getElementById('languageSelector').value;
const langCode = speechLangMap[selectedLang] || 'en-US';
const utterance = new SpeechSynthesisUtterance('Speaker works fine');
let selectedVoice = voices.find(voice => voice.lang === langCode);
if (!selectedVoice) {
selectedVoice = voices.find(voice => voice.lang.startsWith('en')) || voices[0];
statusBar.textContent = `Voice for ${selectedLang} unavailable. Using English voice.`;
}
if (selectedVoice) {
utterance.voice = selectedVoice;
utterance.lang = selectedVoice.lang;
} else {
statusBar.textContent = 'No voices available for text-to-speech.';
return false;
}
utterance.volume = 1.0;
utterance.rate = 1.0;
utterance.pitch = 1.0;
utterance.onerror = (event) => {
console.error('Test TTS error:', event.error);
statusBar.textContent = 'Error testing speaker.';
};
utterance.onend = () => {
console.log('Test TTS finished.');
statusBar.textContent = 'Speaker test successful.';
};
synth.speak(utterance);
return true;
} catch (error) {
console.error('TTS test failed:', error);
statusBar.textContent = 'Failed to test speaker.';
return false;
}
}
// Request microphone permission
async function requestMicPermission() {
try {
const stream = await navigator.mediaDevices.getUserMedia({ audio: true });
stream.getTracks().forEach(track => track.stop());
return true;
} catch (error) {
console.error('Microphone permission error:', error);
statusBar.textContent = 'Microphone access denied. Voice input unavailable.';
return false;
}
}
// Check and request all permissions
async function checkAndRequestPermissions() {
if (browser !== 'Safari') return true;
try {
const permissionStatus = await navigator.permissions.query({ name: 'microphone' });
if (permissionStatus.state === 'granted') {
return await testSpeaker(); // Test speaker if mic is granted
}
} catch (error) {
console.warn('Permission query not supported:', error);
}
permissionModal.style.display = 'flex';
return new Promise((resolve) => {
grantPermissionsButton.onclick = async () => {
permissionModal.style.display = 'none';
const micGranted = await requestMicPermission();
const ttsReady = micGranted ? await testSpeaker() : false;
if (!micGranted || !ttsReady) {
statusBar.textContent = 'Some permissions were not granted. Features may be limited.';
}
resolve(micGranted && ttsReady);
};
});
}
// Speech recognition
function runSpeechRecog() {
const selectedLang = document.getElementById('languageSelector').value;
const action = document.getElementById('action');
if (!window.SpeechRecognition && !window.webkitSpeechRecognition) {
statusBar.textContent = 'Speech recognition not supported in this browser.';
return;
}
const recognition = new (window.SpeechRecognition || window.webkitSpeechRecognition)();
recognition.lang = speechLangMap[selectedLang] || 'en-US';
recognition.interimResults = false;
recognition.continuous = false;
recognition.onstart = () => {
action.textContent = 'Listening...';
statusBar.textContent = '';
};
recognition.onresult = (event) => {
const transcript = event.results[0][0].transcript;
action.textContent = '';
sendMessage(transcript, selectedLang);
};
recognition.onerror = (event) => {
action.textContent = '';
statusBar.textContent = `Speech recognition error: ${event.error}`;
};
recognition.onend = () => {
action.textContent = '';
};
try {
recognition.start();
} catch (error) {
statusBar.textContent = 'Failed to start speech recognition.';
console.error('STT error:', error);
}
}
// Send message to Flask API
async function sendMessage(message, language) {
showUserMessage(message);
try {
const response = await fetch('/api/process_text', {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({ text: message, language })
});
const data = await response.json();
console.log('Response from Flask API:', data);
handleResponse(data);
} catch (error) {
console.error('Error sending data to Flask API:', error);
statusBar.textContent = 'Error: Unable to process request';
showBotMessage('Error: Unable to process request');
}
}
// Handle API response
function handleResponse(data) {
if (data.error) {
statusBar.textContent = data.error;
showBotMessage(data.error);
return;
}
showBotMessage(data.response);
speakResponse(data.response, data.language);
}
// Show user message
function showUserMessage(message) {
const chatBox = document.getElementById('chat-box');
chatBox.innerHTML += `${message}`;
chatBox.scrollTop = chatBox.scrollHeight;
}
// Show bot message
function showBotMessage(message) {
const chatBox = document.getElementById('chat-box');
chatBox.innerHTML += `${message}`;
chatBox.scrollTop = chatBox.scrollHeight;
}
// Initialize
window.addEventListener('load', async () => {
await loadVoices();
if (browser === 'Safari') {
statusBar.textContent = 'Safari detected. Please grant microphone and speaker permissions.';
const permissionsGranted = await checkAndRequestPermissions();
if (!permissionsGranted) {
statusBar.textContent = 'Permissions denied. Some features may not work.';
}
}
document.getElementById('speech').addEventListener('click', runSpeechRecog);
testSpeakerButton.addEventListener('click', testSpeaker);
});
// Clean up on unload
window.addEventListener('beforeunload', () => {
if (synth && synth.speaking) synth.cancel();
});
Подробнее здесь: https://stackoverflow.com/questions/796 ... os-devices
Речь к тексту не работает на устройствах iOS ⇐ Javascript
-
- Похожие темы
- Ответы
- Просмотры
- Последнее сообщение
-
-
TTS в конфликте с Talkback на устройствах Samsung: речь не работает или не прервана
Anonymous » » в форуме Android - 0 Ответы
- 17 Просмотры
-
Последнее сообщение Anonymous
-