Update helpers/tts.js

helpers/tts.js

@@ -33,70 +33,53 @@ if (!key || !region) {
  * @param {*} text text to convert to audio/speech
  * @param {*} filename optional - best for long text - temp file for converted speech/audio
  */
-const textToSpeech = async (text, voice)=> {
-    
-    // convert callback function to promise
-    return new Promise((resolve, reject) => {
-        
-
-        let ssml = SSML.replace("__TEXT__", text);
-
-        
-        const speechConfig = sdk.SpeechConfig.fromSubscription(key, region);
-        speechConfig.speechSynthesisOutputFormat = 5; // mp3
-        
-        let audioConfig = null;
-        
-        // if (filename) {
-        let randomString = Math.random().toString(36).slice(2, 7);
-        let filename = `./public/speech-${randomString}.mp3`;
-        audioConfig = sdk.AudioConfig.fromAudioFileOutput(filename);
-        // }
-
-        let blendData = [];
-        let timeStep = 1/60;
-        let timeStamp = 0;
-
-        const synthesizer = new sdk.SpeechSynthesizer(speechConfig, audioConfig);
-
-        // Subscribes to viseme received event
-        synthesizer.visemeReceived = function (s, e) {
-
-            // `Animation` is an xml string for SVG or a json string for blend shapes
-            var animation = JSON.parse(e.animation);
-
-            _.each(animation.BlendShapes, blendArray => {
-
-                let blend = {};
-                _.each(blendShapeNames, (shapeName, i) => {
-                    blend[shapeName] = blendArray[i];
-                });
-        
-                blendData.push({
-                    time: timeStamp,
-                    blendshapes: blend
-                });
-                console.log(`Timestamp: ${timeStamp.toFixed(3)}s`);
-                console.log(JSON.stringify(blend, null, 2));
-                timeStamp += timeStep;
-            });
-
-        }
-
-
-        synthesizer.speakSsmlAsync(
-            ssml,
-            result => {
-                
-                synthesizer.close();
-                resolve({blendData, filename: `/speech-${randomString}.mp3`});
-
-            },
-            error => {
-                synthesizer.close();
-                reject(error);
-            }); 
-    });
+const textToSpeech = async (text, voice) => {
+  return new Promise((resolve, reject) => {
+    let ssml = SSML.replace("__TEXT__", text);
+
+    const speechConfig = sdk.SpeechConfig.fromSubscription(key, region);
+    speechConfig.speechSynthesisOutputFormat = 5; // mp3
+
+    let audioConfig = null;
+
+    let randomString = Math.random().toString(36).slice(2, 7);
+    let filename = `/tmp/speech-${randomString}.mp3`;  // Write to /tmp directory
+    audioConfig = sdk.AudioConfig.fromAudioFileOutput(filename);
+
+    let blendData = [];
+    let timeStep = 1/60;
+    let timeStamp = 0;
+
+    const synthesizer = new sdk.SpeechSynthesizer(speechConfig, audioConfig);
+
+    synthesizer.visemeReceived = function (s, e) {
+      var animation = JSON.parse(e.animation);
+
+      _.each(animation.BlendShapes, blendArray => {
+        let blend = {};
+        _.each(blendShapeNames, (shapeName, i) => {
+          blend[shapeName] = blendArray[i];
+        });
+
+        blendData.push({
+          time: timeStamp,
+          blendshapes: blend
+        });
+        timeStamp += timeStep;
+      });
+    };
+
+    synthesizer.speakSsmlAsync(
+      ssml,
+      result => {
+        synthesizer.close();
+        resolve({ blendData, filename: `/speech-${randomString}.mp3` });
+      },
+      error => {
+        synthesizer.close();
+        reject(error);
+      }
+    );
+  });
 };
 
-module.exports = textToSpeech;