bible-technology · sijumoncy · Nov 6, 2023 · Nov 9, 2023 · Dec 4, 2023 · Dec 4, 2023
diff --git a/.vscode/settings.json b/.vscode/settings.json
@@ -2,15 +2,11 @@
 	"[javascript]": {
 		"editor.formatOnSave": false
 	},
-	"editor.rulers": [
-		100
-	],
+	"editor.rulers": [100],
 	"editor.fontLigatures": true,
 	"prettier.tabWidth": 4,
 	"eslint.alwaysShowStatus": true,
-	"prettier.disableLanguages": [
-		"js"
-	],
+	"prettier.disableLanguages": ["js"],
 	"prettier.useTabs": true,
 	"editor.formatOnSave": true,
 	"editor.multiCursorModifier": "alt",
@@ -27,8 +23,8 @@
 	"eslint.codeAction.disableRuleComment": {},
 	"eslint.codeAction.showDocumentation": {},
 	"editor.codeActionsOnSave": {
-		"source.fixAll.eslint": true
+		"source.fixAll.eslint": "explicit"
 	},
 	"eslint.workingDirectories": [],
 	"editor.tabSize": 2
-}
+}
diff --git a/package.json b/package.json
@@ -118,6 +118,8 @@
     "@electron/remote": "^2.0.8",
     "@emotion/react": "^11.10.5",
     "@emotion/styled": "^11.10.5",
+    "@ffmpeg/ffmpeg": "^0.12.7",
+    "@ffmpeg/util": "^0.12.1",
     "@headlessui/react": "^1.7.8",
     "@heroicons/react": "^2.0.14",
     "@material-ui/core": "^4.12.4",

diff --git a/renderer/src/components/AudioRecorder/core/audioUtils.js b/renderer/src/components/AudioRecorder/core/audioUtils.js
@@ -1,134 +1,168 @@
 /* eslint-disable no-await-in-loop */
+import { fetchFile } from '@ffmpeg/util';
+import * as localforage from 'localforage';
 import * as logger from '../../../logger';
-
-const toWav = require('audiobuffer-to-wav');
+import packageInfo from '../../../../../package.json';
 
 function sec_to_min_sec_milli_convertor(time) {
-    logger.debug('audioUtils.js', 'In  time conversion function');
-    let milliseconds = time.toString().split('.')[1];
-    if (milliseconds === undefined) {
-        milliseconds = '0';
+    try {
+        logger.debug('audioUtils.js', 'In  time conversion function');
+        let milliseconds = time.toString().split('.')[1];
+        if (milliseconds === undefined) {
+            milliseconds = '0';
+        }
+        const minutes = Math.floor(time / 60);
+        const seconds = (time - minutes * 60).toString().split('.')[0].padStart(2, 0);
+        const formatedStringTime = `${minutes.toString().padStart(2, 0)}:${seconds}:${milliseconds.padStart(2, 0)}`;
+        logger.debug('audioUtils.js', 'In  time conversion function done');
+        return [minutes, seconds, milliseconds, formatedStringTime];
+    } catch (err) {
+        throw new Error(`audio generation failed : ${err}`);
     }
-    const minutes = Math.floor(time / 60);
-    const seconds = (time - minutes * 60).toString().split('.')[0].padStart(2, 0);
-    const formatedStringTime = `${minutes.toString().padStart(2, 0)}:${seconds}:${milliseconds.padStart(2, 0)}`;
-    return [minutes, seconds, milliseconds, formatedStringTime];
 }
 
 async function generateTimeStampData(buffers, book, chapter) {
-    logger.debug('audioUtils.js', 'In TimeStamp Generation');
-    return new Promise((resolve) => {
-        let fileString = 'verse_number\tstart_timestamp\tduration\n';
-        const seperator = '\t';
-        const fileType = 'tsv';
-        const file = `${book}_${chapter.toString().padStart(3, 0)}.${fileType}`;
-        let start = 0;
-        buffers.forEach((buffer, index) => {
-            const currentVerse = `Verse_${(index + 1).toString().padStart(2, 0)}`;
-            const startTimeString = sec_to_min_sec_milli_convertor(start)[3];
-            const durationString = sec_to_min_sec_milli_convertor(buffer.duration)[3];
-            fileString += `${currentVerse + seperator + startTimeString + seperator + durationString}\n`;
-            start += buffer.duration;
+    try {
+        logger.debug('audioUtils.js', 'In TimeStamp Generation');
+        return new Promise((resolve) => {
+            let fileString = 'verse_number\tstart_timestamp\tduration\n';
+            const seperator = '\t';
+            const fileType = 'tsv';
+            const file = `${book}_${chapter.toString().padStart(3, 0)}.${fileType}`;
+            let start = 0; // because of 1 sec silence in merged verses
+            buffers.forEach((buffer, index) => {
+                const currentVerse = `Verse_${(index + 1).toString().padStart(2, 0)}`;
+                const startTimeString = sec_to_min_sec_milli_convertor(start)[3];
+                let durationString;
+                const silenceDuration = 2;
+                let offset = 0;
+                if (index === 0 || (index === buffers.length - 1)) {
+                    // adding this because of 2 sec silence in the start || 2 sec silence in the end add it to last verse
+                    durationString = sec_to_min_sec_milli_convertor(buffer.duration + silenceDuration)[3];
+                    offset = silenceDuration;
+                } else {
+                    offset = 0;
+                    durationString = sec_to_min_sec_milli_convertor(buffer.duration)[3];
+                }
+                fileString += `${currentVerse + seperator + startTimeString + seperator + durationString}\n`;
+                start += buffer.duration + offset;
+            });
+            resolve([file, fileString]);
+        }).catch((err) => {
+            throw new Error(`audio generation failed : ${err}`);
         });
-        resolve([file, fileString]);
-    });
-}
-
-async function fetchAndCombineAudio(audioArr, path) {
-    logger.debug('audioUtils.js', 'In Fetch and merge audio function');
-    // eslint-disable-next-line no-async-promise-executor
-    return new Promise(async (resolve) => {
-        const context = new window.AudioContext();
-
-        // store the decoded buff
-        const sources = [];
-
-        // eslint-disable-next-line no-restricted-syntax, guard-for-in
-        for (const url of audioArr) {
-            try {
-            const response = await fetch(path.join('file://', url));
-            const buffer = await response.arrayBuffer();
-            const decodedData = await context.decodeAudioData(buffer);
-            sources.push(decodedData);
-            } catch (err) {
-                logger.error('audioUtils.js', `Error reading audio - ${url} : ${err}`);
-            }
-        }
-
-        logger.debug('audioUtils.js', 'In fetchAndCombineAudio : Fetch audio success  ');
-
-        const totalLength = sources.reduce((total, source) => total + source.length, 0);
-        const output = context.createBuffer(1, totalLength, context.sampleRate);
-
-        let offset = 0;
-
-        // eslint-disable-next-line no-restricted-syntax
-        for (const source of sources) {
-            output.copyToChannel(source.getChannelData(0), 0, offset);
-            offset += source.length;
-        }
-        const wavData = toWav(output);
-        logger.debug('audioUtils.js', 'In fetchAndCombineAudio : generate wav success');
-        const blob = new Blob([new DataView(wavData)], { type: 'audio/wav' });
-        logger.debug('audioUtils.js', 'In fetchAndCombineAudio : Generate Final merged Audio success ');
-        resolve({ blob, buffers: sources });
-    });
+    } catch (err) {
+        throw new Error('audio generation failed : err');
+    }
 }
 
 function sortingLogic(a, b) {
     // expected format : '1_10_1_default.mp3',
     return a.split('_')[1] - b.split('_')[1];
 }
 
-export async function mergeAudio(audioArr, dirPath, path, book, chapter) {
+export async function mergeAudio(audioArr, dirPath, path, book, chapter, extension, ffmpeg, project) {
+  try {
     logger.debug('audioUtils.js', 'In Merge Audio fucntion');
-    // const audio = new ConcatAudio(window);
-    return new Promise((resolve) => {
+    // eslint-disable-next-line no-async-promise-executor
+    return new Promise(async (resolve, reject) => {
+      try {
         audioArr.sort(sortingLogic);
         for (let i = 0; i < audioArr.length; i++) {
             audioArr[i] = path.join(dirPath, audioArr[i]);
         }
-        logger.debug('audioUtils.js', 'start merging audios');
-
-        fetchAndCombineAudio(audioArr, path)
-        .then(async (mergedData) => {
-            logger.debug('audioUtils.js', `Audio merge success . Started Generate Timestamp : ${book} : ${chapter}`);
-            generateTimeStampData(mergedData.buffers, book, chapter)
-            .then((timeStampData) => {
-                logger.debug('audioUtils.js', `return Merged Audio for chapter : ${book} : ${chapter}`);
-                resolve([mergedData.blob, timeStampData]);
-            });
+        logger.debug('audioUtils.js', 'mergeAudio : start merging audios');
+
+        const inputCmdArr = [];
+        const commandArr = [];
+        const audioBuffersArr = [];
+        let filterStr = '';
+        let currentUser;
+        await localforage.getItem('userProfile').then((value) => {
+            currentUser = value?.username;
         });
+        // only support specific data fields // title, artist, album, year, comment
+        const audioMetaDataArr = [
+            '-metadata', `title=${book}_${chapter}`,
+            '-metadata', `artist=${currentUser || packageInfo.name}`,
+            '-metadata', `album=${project.name}`,
+            '-metadata', `comment=${packageInfo.name}_${packageInfo.version}`,
+            '-metadata', `date=${new Date().getFullYear().toString()}`,
+        ];
+
+        const audioContext = new (window.AudioContext || window.webkitAudioContext)();
+
+        for (let i = 0; i < audioArr.length; i++) {
+          const audioFile = await fetchFile(path.join('file://', audioArr[i]));
+          const audioFileCopy = new Uint8Array(audioFile);
+          await ffmpeg.writeFile(`input_${i}.wav`, audioFile);
+          inputCmdArr.push('-i', `input_${i}.wav`);
+          filterStr += `[${i}:0]`;
+
+          const audioBuffer = await audioContext.decodeAudioData(audioFileCopy.buffer);
+          audioBuffersArr.push(audioBuffer);
+        }
+        filterStr += `concat=n=${audioArr.length + 1 }:v=0:a=1[out]`;
+
+        commandArr.push(
+          ...inputCmdArr,
+          '-filter_complex',
+          `aevalsrc=0:d=2[silence1];[silence1]${filterStr}`,
+          '-map',
+          '[out]',
+          '-ar',
+          '48000',
+          `output.${extension}`,
+        );
+
+        console.log({ filterStr, commandArr });
+
+        logger.debug('audioUtils.js', 'mergeAudio : audio internal write done and buffers created');
+
+        // exeute merge process
+        await ffmpeg.exec(commandArr);
+
+        // add end 2 sec to the create audio
+        await ffmpeg.exec([
+            '-i',
+            `output.${extension}`,
+            '-filter_complex',
+            'aevalsrc=0:d=2[silenceEnd];[0:0][silenceEnd]concat=n=2:v=0:a=1[outEnd]',
+            '-map',
+            '[outEnd]',
+            ...audioMetaDataArr,
+            '-ar',
+            '48000',
+            `outputMerged.${extension}`,
+            ]);
+
+        // write the audio back and convert
+
+        // generate unit8 buffer out
+        const fileData = await ffmpeg.readFile(`outputMerged.${extension}`);
+
+        logger.debug('audioUtils.js', 'mergeAudio : audio merged buffer created');
+
+        // Create a Blob from the result
+        const blob = new Blob([fileData.buffer], { type: extension === 'mp3' ? 'audio/mpeg' : 'audio/wav' });
+        logger.debug('audioUtils.js', 'mergeAudio : audio merged blob created');
+
+        if (blob) {
+          generateTimeStampData(audioBuffersArr, book, chapter)
+          .then((timeStampData) => {
+              logger.debug('audioUtils.js', `mergeAudio : return timestamp for Merged Audio for chapter : ${book} : ${chapter}`);
+              resolve([blob, timeStampData]);
+          }).catch((err) => {
+              throw new Error(`unable to generate audio : ${err}`);
+          });
+        } else {
+            throw new Error('unable to generate audio');
+        }
+      } catch (err) {
+          reject(new Error(`audio generation failed : ${err}`));
+      }
     });
+  } catch (err) {
+      throw new Error(`audio generation failed : ${err}`);
+  }
 }
-
-// old snippet for reference
-
-// export async function mergeAudio(audioArr, dirPath, path, book, chapter) {
-//     logger.debug('audioUtils.js', 'In Merge Audio fucntion');
-//     const audio = new ConcatAudio(window);
-//     return new Promise((resolve) => {
-//         let merged;
-//         let output;
-//         audioArr.sort();
-//         for (let i = 0; i < audioArr.length; i++) {
-//             audioArr[i] = path.join(dirPath, audioArr[i]);
-//         }
-//         logger.debug('audioUtils.js', 'start merging audios');
-//         audio.fetchAudio(...audioArr)
-//         .then(async (buffers) => {
-//             // generate timestamp data string
-//             await generateTimeStampData(buffers, book, chapter)
-//             .then((timeStampData) => {
-//                 // merging all buffers
-//                 merged = audio.concatAudio(buffers);
-//                 return [merged, timeStampData];
-//             })
-//             .then(async ([merged, timeStampData]) => {
-//                 output = audio.export(merged, 'audio/mp3');
-//                 logger.debug('audioUtils.js', `return Merged Audio for chapter : ${book} : ${chapter}`);
-//                 resolve([output.blob, timeStampData]);
-//             });
-//         });
-//     });
-// }
diff --git a/renderer/src/components/EditorPage/AudioEditor/AudioEditor.js b/renderer/src/components/EditorPage/AudioEditor/AudioEditor.js
@@ -158,9 +158,9 @@ const AudioEditor = ({ editor }) => {
                                             } else {
                                               // If found only one audio for the verse then making that audio as default one.
                                               // replace url with `${chapter}_${verseNum[1]}_1_default.mp3`
-                                              bookContent[key].contents[v].take1 = `${chapter}_${verseNum[1]}_1_default.mp3`;
+                                              bookContent[key].contents[v].take1 = `${chapter}_${verseNum[1]}_1_default.wav`;
                                               bookContent[key].contents[v].default = 'take1';
-                                              fs.renameSync(path.join(filePath, chapterNum, verse), path.join(filePath, chapterNum, `${chapter}_${verseNum[1]}_1_default.mp3`));
+                                              fs.renameSync(path.join(filePath, chapterNum, verse), path.join(filePath, chapterNum, `${chapter}_${verseNum[1]}_1_default.wav`));
                                             }
                                           }
                                         },