Reduce SSML size

packages/text-to-speech/src/htmlToSsml.ts

@@ -91,7 +91,7 @@ function emitTextNode(
   if (ssmlElement) {
     emit(textItems, `<${ssmlElement}>`)
   }
-  emit(textItems, `${cleanedText}`)
+  emit(textItems, `${cleanedText.replace(/\s+/g, ' ')}`)
   if (ssmlElement) {
     emit(textItems, `</${ssmlElement}>`)
   }
@@ -123,7 +123,7 @@ function emitElement(
     ) {
       const cleanedText = cleanTextNode(child)
       if (idx && cleanedText.length > 1) {
-        // Make sure its more than just a space
+        // Make sure it's more than just a space
         emit(textItems, `<bookmark mark="${idx}" />`)
       }
       emitTextNode(textItems, cleanedText, child)
@@ -158,9 +158,7 @@ const startSsml = (element: Element, options: SSMLOptions): string => {
     element.nodeName === 'BLOCKQUOTE'
       ? options.secondaryVoice
       : options.primaryVoice
-  return `
-<speak xmlns="http://www.w3.org/2001/10/synthesis" xmlns:mstts="http://www.w3.org/2001/mstts" xmlns:emo="http://www.w3.org/2009/10/emotionml" version="1.0" xml:lang="${options.language}"><voice name="${voice}"><prosody rate="${options.rate}" pitch="default">
-  `
+  return `<speak xmlns="http://www.w3.org/2001/10/synthesis" version="1.0" xml:lang="${options.language}"><voice name="${voice}"><prosody rate="${options.rate}">`
 }
 
 const endSsml = (): string => {

packages/text-to-speech/src/index.ts

@@ -10,7 +10,7 @@ import * as dotenv from 'dotenv' // see https://github.com/motdotla/dotenv#how-d
 import { synthesizeTextToSpeech, TextToSpeechInput } from './textToSpeech'
 import { File, Storage } from '@google-cloud/storage'
 import { PassThrough } from 'stream'
-import { SSMLItem } from './htmlToSsml'
+import { htmlToSsml, SSMLItem } from './htmlToSsml'
 
 dotenv.config()
 Sentry.GCPFunction.init({
@@ -164,3 +164,9 @@ export const textToSpeechStreamingHandler = Sentry.GCPFunction.wrapHttpFunction(
     }
   }
 )
+
+module.exports = {
+  htmlToSsml,
+  textToSpeechStreamingHandler,
+  textToSpeechHandler,
+}

packages/text-to-speech/src/textToSpeech.ts

@@ -46,7 +46,6 @@ export const synthesizeTextToSpeech = async (
     process.env.AZURE_SPEECH_KEY,
     process.env.AZURE_SPEECH_REGION
   )
-  const textType = input.textType || 'html'
   speechConfig.speechSynthesisOutputFormat =
     SpeechSynthesisOutputFormat.Audio16Khz32KBitRateMonoMp3