2019-10-10 23:58:06 +00:00
|
|
|
/*
|
2020-04-10 18:06:55 +00:00
|
|
|
* Copyright (C) 2019-2020 Yomichan Authors
|
2019-10-10 23:58:06 +00:00
|
|
|
*
|
|
|
|
* This program is free software: you can redistribute it and/or modify
|
|
|
|
* it under the terms of the GNU General Public License as published by
|
|
|
|
* the Free Software Foundation, either version 3 of the License, or
|
|
|
|
* (at your option) any later version.
|
|
|
|
*
|
|
|
|
* This program is distributed in the hope that it will be useful,
|
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
* GNU General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU General Public License
|
2020-01-01 17:00:31 +00:00
|
|
|
* along with this program. If not, see <https://www.gnu.org/licenses/>.
|
2019-10-10 23:58:06 +00:00
|
|
|
*/
|
|
|
|
|
2019-10-13 03:59:21 +00:00
|
|
|
class TextToSpeechAudio {
|
|
|
|
constructor(text, voice) {
|
|
|
|
this.text = text;
|
|
|
|
this.voice = voice;
|
|
|
|
this._utterance = null;
|
|
|
|
this._volume = 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
get currentTime() {
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
set currentTime(value) {
|
|
|
|
// NOP
|
|
|
|
}
|
|
|
|
|
|
|
|
get volume() {
|
|
|
|
return this._volume;
|
|
|
|
}
|
|
|
|
set volume(value) {
|
|
|
|
this._volume = value;
|
|
|
|
if (this._utterance !== null) {
|
|
|
|
this._utterance.volume = value;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-04-17 21:48:55 +00:00
|
|
|
async play() {
|
2019-10-13 03:59:21 +00:00
|
|
|
try {
|
|
|
|
if (this._utterance === null) {
|
|
|
|
this._utterance = new SpeechSynthesisUtterance(this.text || '');
|
|
|
|
this._utterance.lang = 'ja-JP';
|
|
|
|
this._utterance.volume = this._volume;
|
|
|
|
this._utterance.voice = this.voice;
|
|
|
|
}
|
|
|
|
|
|
|
|
speechSynthesis.cancel();
|
|
|
|
speechSynthesis.speak(this._utterance);
|
|
|
|
} catch (e) {
|
|
|
|
// NOP
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
pause() {
|
|
|
|
try {
|
|
|
|
speechSynthesis.cancel();
|
|
|
|
} catch (e) {
|
|
|
|
// NOP
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-03-07 17:44:14 +00:00
|
|
|
class AudioSystem {
|
2020-04-10 20:12:55 +00:00
|
|
|
constructor({audioUriBuilder, useCache}) {
|
|
|
|
this._cache = useCache ? new Map() : null;
|
2020-03-07 17:44:14 +00:00
|
|
|
this._cacheSizeMaximum = 32;
|
2020-04-10 17:44:31 +00:00
|
|
|
this._audioUriBuilder = audioUriBuilder;
|
2020-03-07 17:44:14 +00:00
|
|
|
|
|
|
|
if (typeof speechSynthesis !== 'undefined') {
|
|
|
|
// speechSynthesis.getVoices() will not be populated unless some API call is made.
|
|
|
|
speechSynthesis.addEventListener('voiceschanged', this._onVoicesChanged.bind(this));
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-03-07 19:11:28 +00:00
|
|
|
async getDefinitionAudio(definition, sources, details) {
|
|
|
|
const key = `${definition.expression}:${definition.reading}`;
|
2020-04-10 20:12:55 +00:00
|
|
|
const hasCache = (this._cache !== null);
|
|
|
|
|
|
|
|
if (hasCache) {
|
|
|
|
const cacheValue = this._cache.get(key);
|
|
|
|
if (typeof cacheValue !== 'undefined') {
|
|
|
|
const {audio, uri, source} = cacheValue;
|
2020-04-10 20:38:53 +00:00
|
|
|
const index = sources.indexOf(source);
|
|
|
|
if (index >= 0) {
|
|
|
|
return {audio, uri, index};
|
2020-04-10 20:35:43 +00:00
|
|
|
}
|
2020-04-10 20:12:55 +00:00
|
|
|
}
|
2020-03-07 17:44:14 +00:00
|
|
|
}
|
|
|
|
|
2020-04-10 20:38:53 +00:00
|
|
|
for (let i = 0, ii = sources.length; i < ii; ++i) {
|
|
|
|
const source = sources[i];
|
2020-03-07 19:11:28 +00:00
|
|
|
const uri = await this._getAudioUri(definition, source, details);
|
2020-03-07 17:44:14 +00:00
|
|
|
if (uri === null) { continue; }
|
|
|
|
|
|
|
|
try {
|
2020-04-10 17:44:31 +00:00
|
|
|
const audio = await this._createAudio(uri);
|
2020-04-10 20:12:55 +00:00
|
|
|
if (hasCache) {
|
|
|
|
this._cacheCheck();
|
|
|
|
this._cache.set(key, {audio, uri, source});
|
|
|
|
}
|
2020-04-10 20:38:53 +00:00
|
|
|
return {audio, uri, index: i};
|
2020-03-07 17:44:14 +00:00
|
|
|
} catch (e) {
|
|
|
|
// NOP
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
throw new Error('Could not create audio');
|
|
|
|
}
|
|
|
|
|
2020-04-10 17:51:47 +00:00
|
|
|
createTextToSpeechAudio(text, voiceUri) {
|
2020-03-07 17:44:14 +00:00
|
|
|
const voice = this._getTextToSpeechVoiceFromVoiceUri(voiceUri);
|
|
|
|
if (voice === null) {
|
|
|
|
throw new Error('Invalid text-to-speech voice');
|
|
|
|
}
|
|
|
|
return new TextToSpeechAudio(text, voice);
|
|
|
|
}
|
|
|
|
|
|
|
|
_onVoicesChanged() {
|
|
|
|
// NOP
|
|
|
|
}
|
|
|
|
|
2020-04-10 17:44:31 +00:00
|
|
|
async _createAudio(uri) {
|
2020-03-07 17:44:14 +00:00
|
|
|
const ttsParameters = this._getTextToSpeechParameters(uri);
|
|
|
|
if (ttsParameters !== null) {
|
2020-04-10 17:51:47 +00:00
|
|
|
const {text, voiceUri} = ttsParameters;
|
|
|
|
return this.createTextToSpeechAudio(text, voiceUri);
|
2020-03-07 17:44:14 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
return await this._createAudioFromUrl(uri);
|
|
|
|
}
|
|
|
|
|
2020-04-10 17:44:31 +00:00
|
|
|
_getAudioUri(definition, source, details) {
|
|
|
|
return (
|
|
|
|
this._audioUriBuilder !== null ?
|
|
|
|
this._audioUriBuilder.getUri(definition, source, details) :
|
|
|
|
null
|
|
|
|
);
|
|
|
|
}
|
|
|
|
|
2020-03-07 17:44:14 +00:00
|
|
|
_createAudioFromUrl(url) {
|
|
|
|
return new Promise((resolve, reject) => {
|
|
|
|
const audio = new Audio(url);
|
|
|
|
audio.addEventListener('loadeddata', () => {
|
|
|
|
const duration = audio.duration;
|
|
|
|
if (duration === 5.694694 || duration === 5.720718) {
|
|
|
|
// Hardcoded values for invalid audio
|
|
|
|
reject(new Error('Could not retrieve audio'));
|
|
|
|
} else {
|
|
|
|
resolve(audio);
|
|
|
|
}
|
|
|
|
});
|
|
|
|
audio.addEventListener('error', () => reject(audio.error));
|
|
|
|
});
|
|
|
|
}
|
|
|
|
|
|
|
|
_getTextToSpeechVoiceFromVoiceUri(voiceUri) {
|
|
|
|
try {
|
|
|
|
for (const voice of speechSynthesis.getVoices()) {
|
|
|
|
if (voice.voiceURI === voiceUri) {
|
|
|
|
return voice;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
} catch (e) {
|
|
|
|
// NOP
|
|
|
|
}
|
|
|
|
return null;
|
|
|
|
}
|
|
|
|
|
|
|
|
_getTextToSpeechParameters(uri) {
|
|
|
|
const m = /^tts:[^#?]*\?([^#]*)/.exec(uri);
|
|
|
|
if (m === null) { return null; }
|
|
|
|
|
|
|
|
const searchParameters = new URLSearchParams(m[1]);
|
|
|
|
const text = searchParameters.get('text');
|
|
|
|
const voiceUri = searchParameters.get('voice');
|
|
|
|
return (text !== null && voiceUri !== null ? {text, voiceUri} : null);
|
|
|
|
}
|
|
|
|
|
|
|
|
_cacheCheck() {
|
|
|
|
const removeCount = this._cache.size - this._cacheSizeMaximum;
|
|
|
|
if (removeCount <= 0) { return; }
|
|
|
|
|
|
|
|
const removeKeys = [];
|
|
|
|
for (const key of this._cache.keys()) {
|
|
|
|
removeKeys.push(key);
|
|
|
|
if (removeKeys.length >= removeCount) { break; }
|
|
|
|
}
|
|
|
|
|
|
|
|
for (const key of removeKeys) {
|
|
|
|
this._cache.delete(key);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|