echogarden 2.1.2 → 2.2.0

package/src/synthesis/KokoroTTS.ts

@@ -111,7 +111,7 @@ export class KokoroTTS {
 
 		logger.end()
 
-		const clauseBreakTokenId = charToTokenIDLookup[',']
+		const phraseBreakTokenId = charToTokenIDLookup[',']
 		const wordBreakTokenId = charToTokenIDLookup[' ']
 
 		let sentenceEndChar: string
@@ -128,11 +128,11 @@ export class KokoroTTS {
 
 		const allTokenIds: number[] = []
 
-		for (let clauseIndex = 0; clauseIndex < phonemizedSentence.length; clauseIndex++) {
-			const clause = phonemizedSentence[clauseIndex]
+		for (let phraseIndex = 0; phraseIndex < phonemizedSentence.length; phraseIndex++) {
+			const phrase = phonemizedSentence[phraseIndex]
 
-			for (let wordIndex = 0; wordIndex < clause.length; wordIndex++) {
-				const word = clause[wordIndex]
+			for (let wordIndex = 0; wordIndex < phrase.length; wordIndex++) {
+				const word = phrase[wordIndex]
 
 				for (const phoneme of word) {
 					let processedPhoneme = phoneme
@@ -154,11 +154,11 @@ export class KokoroTTS {
 					}
 				}
 
-				if (wordIndex < clause.length - 1) {
+				if (wordIndex < phrase.length - 1) {
 					allTokenIds.push(wordBreakTokenId)
 				} else {
-					if (clauseIndex < phonemizedSentence.length - 1) {
-						allTokenIds.push(clauseBreakTokenId)
+					if (phraseIndex < phonemizedSentence.length - 1) {
+						allTokenIds.push(phraseBreakTokenId)
 						allTokenIds.push(wordBreakTokenId)
 					}
 				}
@@ -181,10 +181,10 @@ export class KokoroTTS {
 				if (endIndex >= allTokenIds.length) {
 					endIndex = allTokenIds.length
 				} else {
-					const indexOfLastClauseBreak = indexOfLastMatchingNumberInRange(allTokenIds, clauseBreakTokenId, startIndex, endIndex)
+					const indexOfLastPhraseBreak = indexOfLastMatchingNumberInRange(allTokenIds, phraseBreakTokenId, startIndex, endIndex)
 
-					if (indexOfLastClauseBreak !== -1) {
-						endIndex = indexOfLastClauseBreak + 1
+					if (indexOfLastPhraseBreak !== -1) {
+						endIndex = indexOfLastPhraseBreak + 1
 					} else {
 						const indexOfLastWordBreak = indexOfLastMatchingNumberInRange(allTokenIds, wordBreakTokenId, startIndex, endIndex)
 
@@ -196,14 +196,14 @@ export class KokoroTTS {
 
 				const partTokenIds = allTokenIds.slice(startIndex, endIndex)
 
-				if (partTokenIds[partTokenIds.length - 1] === clauseBreakTokenId) {
+				if (partTokenIds[partTokenIds.length - 1] === phraseBreakTokenId) {
 					partTokenIds.pop()
 				}
 
 
 				if (partTokenIds.find(
-					tokenId => ![wordBreakTokenId, clauseBreakTokenId, sentenceEndTokenId].includes(tokenId))) {
-						
+					tokenId => ![wordBreakTokenId, phraseBreakTokenId, sentenceEndTokenId].includes(tokenId))) {
+
 					parts.push([0, ...partTokenIds, 0])
 				}
 
@@ -252,7 +252,7 @@ export class KokoroTTS {
 
 		const { alignUsingDtw } = await import('../alignment/SpeechAlignment.js')
 
-		const referenceWordTimeline = referenceTimeline.flatMap(clause => clause.timeline!)
+		const referenceWordTimeline = referenceTimeline.flatMap(phrase => phrase.timeline!)
 
 		const dtwWindowDuration = Math.max(5, Math.ceil(0.2 * getRawAudioDuration(synthesizedAudio)))
 		const mappedTimeline = await alignUsingDtw(synthesizedAudio, referenceSynthesizedAudio, referenceWordTimeline, ['high'], [dtwWindowDuration])

package/src/synthesis/VitsTTS.ts

@@ -94,7 +94,7 @@ export class VitsTTS {
 
 		await logger.startAsync('Encode phonemes to identifiers')
 
-		const clauseEndBreaker = ','
+		const phraseEndBreaker = ','
 		let sentenceEndBreaker = '.'
 
 		if (sentence.endsWith('?') || sentence.endsWith(`?"`)) {
@@ -108,15 +108,15 @@ export class VitsTTS {
 		const startId = phonemeMap.get('^')!
 		const endId = phonemeMap.get('$')!
 
-		const clauseEndBreakerId = phonemeMap.get(clauseEndBreaker)!
+		const phraseEndBreakerId = phonemeMap.get(phraseEndBreaker)!
 		const sentenceEndBreakerId = phonemeMap.get(sentenceEndBreaker)!
 
 		const ids: number[] = [...startId, ...phonemeCharacterSeparatorId]
 
-		for (let clauseIndex = 0; clauseIndex < phonemizedSentence.length; clauseIndex++) {
-			const clause = phonemizedSentence[clauseIndex]
+		for (let phraseIndex = 0; phraseIndex < phonemizedSentence.length; phraseIndex++) {
+			const phrase = phonemizedSentence[phraseIndex]
 
-			for (const word of clause) {
+			for (const word of phrase) {
 				for (const phoneme of word) {
 					for (const phonemeCharacter of phoneme) {
 						const id = phonemeMap.get(phonemeCharacter)
@@ -130,13 +130,13 @@ export class VitsTTS {
 					}
 				}
 
-				if (clauseIndex < phonemizedSentence.length - 1) {
+				if (phraseIndex < phonemizedSentence.length - 1) {
 					ids.push(...wordSeparatorId, ...phonemeCharacterSeparatorId)
 				}
 			}
 
-			if (clauseIndex < phonemizedSentence.length - 1) {
-				ids.push(...clauseEndBreakerId, ...phonemeCharacterSeparatorId)
+			if (phraseIndex < phonemizedSentence.length - 1) {
+				ids.push(...phraseEndBreakerId, ...phonemeCharacterSeparatorId)
 			}
 		}
 
@@ -174,7 +174,7 @@ export class VitsTTS {
 
 		const { alignUsingDtw } = await import('../alignment/SpeechAlignment.js')
 
-		const referenceWordTimeline = referenceTimeline.flatMap(clause => clause.timeline!)
+		const referenceWordTimeline = referenceTimeline.flatMap(phrase => phrase.timeline!)
 
 		const dtwWindowDuration = Math.max(5, Math.ceil(0.2 * getRawAudioDuration(synthesizedAudio)))
 		const mappedTimeline = await alignUsingDtw(synthesizedAudio, referenceSynthesizedAudio, referenceWordTimeline, ['high'], [dtwWindowDuration])

package/src/utilities/Utilities.ts

@@ -505,3 +505,9 @@ export function indexOfLastMatchingNumberInRange(values: number[], targetValue:
 
 	return -1
 }
+
+export function encodeHTMLAngleBrackets(text: string) {
+	return text.replaceAll('<', '&lt;')
+		.replaceAll('>', '&gt;')
+		.replaceAll('&', '&amp;')
+}