Add dotenv

packages/api/src/routers/article_router.ts

@@ -153,7 +153,14 @@ export function articleRouter() {
         voice: voice || userPersonalization?.speechVoice || 'en-US-JennyNeural',
       })
       // enqueue a task to convert text to speech
-      const taskName = await enqueueTextToSpeech(uid, speech.id)
+      const taskName = await enqueueTextToSpeech(
+        uid,
+        speech.id,
+        page.content,
+        'ssml',
+        speech.voice,
+        env.fileUpload.gcsUploadBucket
+      )
       logger.info('Start Text to speech task', { taskName })
       res.status(202).send('Text to speech task started')
     }

packages/api/src/utils/createTask.ts

@@ -330,12 +330,19 @@ export const enqueueSyncWithIntegration = async (
 
 export const enqueueTextToSpeech = async (
   userId: string,
-  speechId: string
+  speechId: string,
+  text: string,
+  textType: 'text' | 'ssml',
+  voice: string,
+  bucket: string
 ): Promise<string> => {
   const { GOOGLE_CLOUD_PROJECT } = process.env
   const payload = {
-    userId,
-    speechId,
+    id: speechId,
+    text,
+    voice,
+    bucket,
+    textType,
   }
   // eslint-disable-next-line @typescript-eslint/ban-ts-comment
   // @ts-ignore

packages/text-to-speech/package.json

@@ -28,6 +28,7 @@
     "@google-cloud/storage": "^6.4.1",
     "@sentry/serverless": "^6.16.1",
     "axios": "^0.27.2",
+    "dotenv": "^16.0.1",
     "jsonwebtoken": "^8.5.1",
     "linkedom": "^0.14.12",
     "microsoft-cognitiveservices-speech-sdk": "^1.22.0"

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

@@ -17,6 +17,8 @@ import {
 } from 'microsoft-cognitiveservices-speech-sdk'
 import axios from 'axios'
 import * as jwt from 'jsonwebtoken'
+import * as dotenv from 'dotenv' // see https://github.com/motdotla/dotenv#how-do-i-use-dotenv-with-import
+dotenv.config()
 
 interface TextToSpeechInput {
   id: string
@@ -60,9 +62,10 @@ const createGCSFile = (bucket: string, filename: string): File => {
 
 const updateSpeech = async (
   speechId: string,
-  audioFileName: string,
-  speechMarksFileName: string,
-  token: string
+  token: string,
+  state: 'COMPLETED' | 'FAILED',
+  audioFileName?: string,
+  speechMarksFileName?: string
 ): Promise<boolean> => {
   if (!process.env.REST_BACKEND_ENDPOINT) {
     throw new Error('backend rest api endpoint not exists')
@@ -73,6 +76,7 @@ const updateSpeech = async (
       speechId,
       audioFileName,
       speechMarksFileName,
+      state,
     }
   )
 
@@ -82,7 +86,7 @@ const updateSpeech = async (
 const synthesizeTextToSpeech = async (
   input: TextToSpeechInput
 ): Promise<TextToSpeechOutput> => {
-  if (!process.env.azureSpeechKey || !process.env.azureSpeechRegion) {
+  if (!process.env.AZURE_SPEECH_KEY || !process.env.AZURE_SPEECH_REGION) {
     throw new Error('Azure Speech Key or Region not set')
   }
   const audioFileName = `speech/${input.id}.mp3`
@@ -91,8 +95,8 @@ const synthesizeTextToSpeech = async (
     resumable: true,
   })
   const speechConfig = SpeechConfig.fromSubscription(
-    process.env.azureSpeechKey,
-    process.env.azureSpeechRegion
+    process.env.AZURE_SPEECH_KEY,
+    process.env.AZURE_SPEECH_REGION
   )
   const textType = input.textType || 'text'
   if (textType === 'text') {
@@ -393,19 +397,26 @@ export const textToSpeechHandler = Sentry.GCPFunction.wrapHttpFunction(
       return res.status(200).send('UNAUTHENTICATED')
     }
     const input = req.body as TextToSpeechInput
-    const { audioFileName, speechMarksFileName } = await synthesizeTextToSpeech(
-      input
-    )
-    const updated = await updateSpeech(
-      input.id,
-      audioFileName,
-      speechMarksFileName,
-      token
-    )
+    try {
+      const { audioFileName, speechMarksFileName } =
+        await synthesizeTextToSpeech(input)
+      const updated = await updateSpeech(
+        input.id,
+        token,
+        'COMPLETED',
+        audioFileName,
+        speechMarksFileName
+      )
 
-    if (!updated) {
-      return res.status(500).send('Failed to update speech')
+      if (!updated) {
+        return res.status(500).send('Failed to update speech')
+      }
+    } catch (e) {
+      console.error(e)
+      await updateSpeech(input.id, token, 'FAILED')
+      return res.status(500).send('Failed to synthesize')
     }
+
     res.send('OK')
   }
 )

yarn.lock

@@ -12474,6 +12474,11 @@ dotenv@^16.0.0:
   resolved "https://registry.yarnpkg.com/dotenv/-/dotenv-16.0.0.tgz#c619001253be89ebb638d027b609c75c26e47411"
   integrity sha512-qD9WU0MPM4SWLPJy/r2Be+2WgQj8plChsyrCNQzW/0WjvcJQiKQJ9mH3ZgB3fxbUUxgc/11ZJ0Fi5KiimWGz2Q==
 
+dotenv@^16.0.1:
+  version "16.0.1"
+  resolved "https://registry.yarnpkg.com/dotenv/-/dotenv-16.0.1.tgz#8f8f9d94876c35dac989876a5d3a82a267fdce1d"
+  integrity sha512-1K6hR6wtk2FviQ4kEiSjFiH5rpzEVi8WW0x96aztHVMhEspNpc4DVOUTEHtEva5VThQ8IaBX1Pe4gSzpVVUsKQ==
+
 dotenv@^8.0.0, dotenv@^8.2.0:
   version "8.6.0"
   resolved "https://registry.yarnpkg.com/dotenv/-/dotenv-8.6.0.tgz#061af664d19f7f4d8fc6e4ff9b584ce237adcb8b"