Summarizer und weiters angepasst

New Folder structure

See merge request proj-wise2526-video2document/video2document!21

.gitlab-ci.yml

@@ -1,6 +1,33 @@
-build-job:
+workflow:
-    script:
+  rules:
-        - echo "Building the Project.."
+    # Run the pipeline for merge requests or when committing to a branch
-test-job:
+    - if: $CI_PIPELINE_SOURCE == "merge_request_event"
-    script:
+    - if: $CI_COMMIT_BRANCH
-        - echo "Running Tests.."        
+
+image: python:3.14.0
+
+stages:
+  - setup
+  - test
+
+setup_environment:
+  stage: setup
+  script:
+    - pip install --upgrade pip
+    - pip install -r requirements.txt
+    - echo "Dependencies installed successfully."
+
+  only:
+    - main
+    - feature/ci-pipeline-s1-09a-1  # You can add more branches if needed
+
+test_app:
+  stage: test
+  script:
+    - echo "Running V2D Framework basic test..."
+    - python -m unittest discover || echo "No tests found."
+
+  only:
+    - main
+    - feature/ci-pipeline-s1-09a-1
+

app/main.py

@@ -0,0 +1,7 @@
+from fastapi import FastAPI
+
+app = FastAPI()
+
+@app.get("/health")
+def health_check():
+    return {"status": "ok"}

electron/main/preload.js

@@ -12,6 +12,9 @@ try {
     contextBridge.exposeInMainWorld("electronAPI", {
         getFilePath: (file) => {return webUtils.getPathForFile(file)}
     })
+    contextBridge.exposeInMainWorld("summarizer", {
+        runFile: (file) => ipcRenderer.send("summarize-transcription", file)
+    });
 } catch (error) {
     console.log("Error in preload.js");
 }

electron/main/renderer.js

@@ -16,15 +16,21 @@ uploadContainer.addEventListener("drop", (e) => {
         e.preventDefault()
         const files = e.dataTransfer.files
         const filePath = window.explorer.onFileDrop(files[0])
-        var holdy = filePath + "";
-        if(holdy.endsWith(".mp4") || holdy.endsWith(".mov") || holdy.endsWith(".avi") || holdy.endsWith( ".mkv")){
-            console.log(filePath)
 
+        var holdy = String(filePath);
+        const lower = holdy.toLowerCase();
+        const validExt = [".mp4", ".mov", ".avi", ".mkv"];
+        
+       if(validExt.some(ext => lower.endsWith(ext))){
+            console.log(filePath);
             const files1 = e.dataTransfer.files;
             handleFiles(files1);
+        }else{
+            console.log('Video format invalid!');
         }
     } catch (error) {
         console.log("Error in renderer.js with the listerner for the drop function");
+        console.log(error);
     }
     
     

electron/main/script.js

@@ -15,18 +15,24 @@ function checkBoxes() {
     const checkboxes = document.querySelectorAll('input[name="docFormat"]');
     let isChecked = false;
 
-    checkboxes.forEach(function(checkbox){
+    checkboxes.forEach(function (checkbox) {
-      if(checkbox.checked){
+      if (checkbox.checked) {
         isChecked = true;
       }
     });
 
-    if(isChecked){
+    if (isChecked) {
       //Code to submit the video
       var pathTest = window.electronAPI.getFilePath(videoUpload.files[0]);
-      if(pathTest.endsWith(".mp4") || holdy.endsWith(".mov") || holdy.endsWith(".avi") || holdy.endsWith( ".mkv")){
+
-        window.extractor.extract({inputVideoPath: pathTest, outputType:"wav"})
+      const lower = pathTest.toLowerCase();
+      const validExt = [".mp4", ".mov", ".avi", ".mkv"];
+
+      if(validExt.some(ext => lower.endsWith(ext))){
+        window.extractor.extract({ inputVideoPath: pathTest, outputType: "wav" });
       }
+
+
     } else {
       //language only english at the moment
       alert('Please select at least one document type.');
@@ -35,42 +41,42 @@ function checkBoxes() {
     console.log(error)
   }
 
-// mapFunctions.get("extraction-video-to-audio").function({inputVideoPath:"./a.mp4", outputType:"wav"})
+  // mapFunctions.get("extraction-video-to-audio").function({inputVideoPath:"./a.mp4", outputType:"wav"})
 }
 
 //language changing feature
 function changeLanguage(language) {
-    if (language === 'en') {
+  if (language === 'en') {
-        document.getElementById('title').textContent = 'Video to document';
+    document.getElementById('title').textContent = 'Video to document';
-        document.getElementById('h1').textContent = 'Video to document';
+    document.getElementById('h1').textContent = 'Video to document';
-        document.getElementById('p1').textContent = 'Drag and drop video file';
+    document.getElementById('p1').textContent = 'Drag and drop video file';
-        document.getElementById('fileName').textContent = 'No video chosen';
+    document.getElementById('fileName').textContent = 'No video chosen';
-        document.getElementById('manualUploadBtn').textContent = 'Search video';
+    document.getElementById('manualUploadBtn').textContent = 'Search video';
-        document.getElementById('checkbox_group').textContent = 'Choose prefered document style:';
+    document.getElementById('checkbox_group').textContent = 'Choose prefered document style:';
-        document.getElementById('label_format').textContent = 'Meeting report';
+    document.getElementById('label_format').textContent = 'Meeting report';
-        document.getElementById('label_summary').textContent = 'Summary with timestamps';
+    document.getElementById('label_summary').textContent = 'Summary with timestamps';
-        document.getElementById('submitButton').textContent = 'Submit';
+    document.getElementById('submitButton').textContent = 'Submit';
-    } else if (language === 'de') {
+  } else if (language === 'de') {
-        document.getElementById('title').textContent = 'Video zu Dokument';
+    document.getElementById('title').textContent = 'Video zu Dokument';
-        document.getElementById('h1').textContent = 'Video zu Dokument';
+    document.getElementById('h1').textContent = 'Video zu Dokument';
-        document.getElementById('p1').textContent = 'Video per Drag & Drop ablegen';
+    document.getElementById('p1').textContent = 'Video per Drag & Drop ablegen';
-        document.getElementById('fileName').textContent = 'Kein Video ausgewaehlt';
+    document.getElementById('fileName').textContent = 'Kein Video ausgewaehlt';
-        document.getElementById('manualUploadBtn').textContent = 'Video suchen';
+    document.getElementById('manualUploadBtn').textContent = 'Video suchen';
-        document.getElementById('checkbox_group').textContent = 'Bevorzugte Dokumentvarianten:';
+    document.getElementById('checkbox_group').textContent = 'Bevorzugte Dokumentvarianten:';
-        document.getElementById('label_format').textContent = 'Meeting Bericht';
+    document.getElementById('label_format').textContent = 'Meeting Bericht';
-        document.getElementById('label_summary').textContent = 'Zusammenfassung mit Zeitstempeln';
+    document.getElementById('label_summary').textContent = 'Zusammenfassung mit Zeitstempeln';
-        document.getElementById('submitButton').textContent = 'Absenden';
+    document.getElementById('submitButton').textContent = 'Absenden';
-    } else if(language == "in") {
+  } else if (language == "in") {
-        document.getElementById('title').textContent = 'दस्तावेज़ के लिए वीडियो';
+    document.getElementById('title').textContent = 'दस्तावेज़ के लिए वीडियो';
-        document.getElementById('h1').textContent = 'दस्तावेज़ के लिए वीडियो';
+    document.getElementById('h1').textContent = 'दस्तावेज़ के लिए वीडियो';
-        document.getElementById('p1').textContent = 'वीडियो फ़ाइल खींचें और छोड़ें';
+    document.getElementById('p1').textContent = 'वीडियो फ़ाइल खींचें और छोड़ें';
-        document.getElementById('fileName').textContent = 'कोई वीडियो नहीं चुना गया';
+    document.getElementById('fileName').textContent = 'कोई वीडियो नहीं चुना गया';
-        document.getElementById('manualUploadBtn').textContent = 'वीडियो खोजें';
+    document.getElementById('manualUploadBtn').textContent = 'वीडियो खोजें';
-        document.getElementById('checkbox_group').textContent = 'पसंदीदा दस्तावेज़ शैली चुनें:';
+    document.getElementById('checkbox_group').textContent = 'पसंदीदा दस्तावेज़ शैली चुनें:';
-        document.getElementById('label_format').textContent = 'बैठक रिपोर्ट';
+    document.getElementById('label_format').textContent = 'बैठक रिपोर्ट';
-        document.getElementById('label_summary').textContent = 'टाइमस्टैम्प के साथ सारांश';
+    document.getElementById('label_summary').textContent = 'टाइमस्टैम्प के साथ सारांश';
-        document.getElementById('submitButton').textContent = 'जमा करना';
+    document.getElementById('submitButton').textContent = 'जमा करना';
-    }
+  }
 }
 
 
@@ -103,7 +109,7 @@ function handleFiles(files) {
 }
 
 //function to regulate the progress on the progressbar
-function updateProgressBar(bar, value){
+function updateProgressBar(bar, value) {
   try {
     value = Math.round(value);
     bar.querySelector(".progress_fill").style.width = `${value}%`;

main.js

@@ -77,3 +77,8 @@ electron.app.whenReady().then(createWindow);
 electron.ipcMain.on("extract", (event, args) => {
     mapFunctions.get("extraction-video-to-audio").function(args)
 })
+
+electron.ipcMain.on("summarize-transcription", (event, args) => {
+    mapFunctions.get("summarize-transcription").function(args);
+});
+

package-lock.json

@@ -9,7 +9,9 @@
       "version": "1.0.0",
       "license": "ISC",
       "dependencies": {
+        "@types/axios": "^0.9.36",
         "cli-progress": "^3.12.0",
+        "dotenv": "^17.2.3",
         "electron": "^39.1.1",
         "express": "^5.1.0",
         "ffmpeg-static": "^5.2.0",
@@ -149,6 +151,12 @@
       "dev": true,
       "license": "MIT"
     },
+    "node_modules/@types/axios": {
+      "version": "0.9.36",
+      "resolved": "https://registry.npmjs.org/@types/axios/-/axios-0.9.36.tgz",
+      "integrity": "sha512-NLOpedx9o+rxo/X5ChbdiX6mS1atE4WHmEEIcR9NLenRVa5HoVjAvjafwU3FPTqnZEstpoqCaW7fagqSoTDNeg==",
+      "license": "MIT"
+    },
     "node_modules/@types/cacheable-request": {
       "version": "6.0.3",
       "resolved": "https://registry.npmjs.org/@types/cacheable-request/-/cacheable-request-6.0.3.tgz",
@@ -198,6 +206,7 @@
       "resolved": "https://registry.npmjs.org/@types/node/-/node-24.9.2.tgz",
       "integrity": "sha512-uWN8YqxXxqFMX2RqGOrumsKeti4LlmIMIyV0lgut4jx7KQBcBiW6vkDtIBvHnHIquwNfJhk8v2OtmO8zXWHfPA==",
       "license": "MIT",
+      "peer": true,
       "dependencies": {
         "undici-types": "~7.16.0"
       }
@@ -584,6 +593,18 @@
         "node": ">=0.3.1"
       }
     },
+    "node_modules/dotenv": {
+      "version": "17.2.3",
+      "resolved": "https://registry.npmjs.org/dotenv/-/dotenv-17.2.3.tgz",
+      "integrity": "sha512-JVUnt+DUIzu87TABbhPmNfVdBDt18BLOWjMUFJMSi/Qqg7NTYtabbvSNJGOJ7afbRuv9D/lngizHtP7QyLQ+9w==",
+      "license": "BSD-2-Clause",
+      "engines": {
+        "node": ">=12"
+      },
+      "funding": {
+        "url": "https://dotenvx.com"
+      }
+    },
     "node_modules/dunder-proto": {
       "version": "1.0.1",
       "resolved": "https://registry.npmjs.org/dunder-proto/-/dunder-proto-1.0.1.tgz",
@@ -1824,6 +1845,7 @@
       "integrity": "sha512-jl1vZzPDinLr9eUt3J/t7V6FgNEw9QjvBPdysz9KfQDD41fQrC2Y4vKQdiaUpFT4bXlb1RHhLpp8wtm6M5TgSw==",
       "dev": true,
       "license": "Apache-2.0",
+      "peer": true,
       "bin": {
         "tsc": "bin/tsc",
         "tsserver": "bin/tsserver"

package.json

@@ -1,6 +1,8 @@
 {
   "dependencies": {
+    "@types/axios": "^0.9.36",
     "cli-progress": "^3.12.0",
+    "dotenv": "^17.2.3",
     "electron": "^39.1.1",
     "express": "^5.1.0",
     "ffmpeg-static": "^5.2.0",

requirements.txt

@@ -0,0 +1,3 @@
+fastapi
+uvicorn
+pytest

scripts/commands.txt

@@ -1,5 +0,0 @@
-npx ts-node ./extract.ts /Users/mikehughes/Downloads/Testvideo/Kurzgesagt.mov
-npx ts-node ./transcribe.ts ../storage/audio/Kurzgesagt.wav
-
-npx ts-node ./extract.ts /Users/mikehughes/Downloads/Testvideo/GitLabMeeting.mov
-npx ts-node ./transcribe.ts ../storage/audio/GitLabMeeting.wav

scripts/extract.ts

@@ -1,23 +0,0 @@
-#!/usr/bin/env ts-node
-
-import { extractAudioFromVideo } from "../services/modules/extraction/ffmpegExtractor.ts";
-
-const videoPath = process.argv[2];
-
-if (!videoPath) {
-  console.error("Usage: ts-node extractAudio.ts <videoPath>");
-  process.exit(1);
-}
-
-(async () => {
-  try {
-    console.log(`Extracting audio from: ${videoPath}`);
-
-    await extractAudioFromVideo(videoPath); // Call the extraction function (ffmpegExtractor.ts in services/modules/extraction)
-
-    console.log("Audio extraction completed successfully.");
-  } catch (err) {
-    console.error("Audio extraction failed:", err);
-    process.exit(1);
-  }
-})();

scripts/transcribe.ts

@@ -1,18 +0,0 @@
-import { whisperLocal } from "../services/modules/transcription/local/whisperLocal.ts";
-
-const audioPath = process.argv[2];
-if (!audioPath) {
-  console.error("Please provide an audio file path as argument.");
-  process.exit(1);
-}
-
-const whisper = new whisperLocal();
-
-(async () => {
-  try {
-    const text = await whisper.transcribe(audioPath);
-    console.log(text);
-  } catch (err) {
-    console.error("Transcription failed:", err);
-  }
-})();

services/modules/jsonTools/transcriptionSummarizer.js

@@ -0,0 +1,138 @@
+const fs = require("fs");
+const path = require("path");
+
+// Prepare output directory (always storage/transcriptionSummaries under project root)
+const outputDir = `${__dirname}/../../../storage/transcriptionSummaries`;
+if (!fs.existsSync(outputDir)) {
+    fs.mkdirSync(outputDir, { recursive: true });
+}
+
+//Speaker, ALL-Sentences, Start, End
+
+module.exports = {
+    name: "summarize-transcription", // Unique name for our function that will later be used to get the function from the map via "mapFunctions.get("example").function()"
+    type: "summarizer", // value used to differentiate each module to order them in the UI
+    displayname: "Summarizer",  // The displayname used within the UI
+    async function(args) {
+        let inputJson = args.json;
+
+        //JSON Path
+        if (args.jsonPath) {
+            try {
+                const raw = fs.readFileSync(args.jsonPath, "utf-8");
+                inputJson = JSON.parse(raw);
+            } catch (e) {
+                console.error("Failed to load JSON from file:", e);
+                return { error: "Could not read JSON from file path." };
+            }
+        }
+        // JSON parsen
+        if (typeof inputJson === "string") {
+            try {
+                inputJson = JSON.parse(inputJson);
+            } catch (e) {
+                console.log("Invalid JSON in summarize-transcription");
+                return { error: "Invalid JSON" };
+            }
+        }
+
+        const words = inputJson.words;
+        if (!Array.isArray(words)) {
+            return { error: "No words Array found" };
+        }
+
+        const ENDINGS = [".", "!", "?"]; // '...' auch als Satzende ?
+        const ABBREVIATIONS = new Set(["z.B.", "bzw.", "u.a.", "Dr.", "Mr.", "Mrs.", "Prof.", "etc."]); //TODO weitere Ergaenzen
+
+        const result = [];
+        let currentSentence = "";
+        let currentSpeaker = null;
+        let startTime = null;
+        let endTime = null;
+
+        for (const w of words) {
+            if (!currentSpeaker) currentSpeaker = w.speaker;
+            if (startTime === null) startTime = w.start;
+            endTime = w.end;
+    
+            //speaker changing
+            if (currentSpeaker !== w.speaker && currentSentence) { 
+                const lastEntry = result[result.length - 1];
+                if (lastEntry && lastEntry.speaker === currentSpeaker) {
+                    lastEntry.sentence += " " + currentSentence;
+                    lastEntry.end = endTime;
+                } else {
+                    result.push({
+                        speaker: currentSpeaker,
+                        sentence: currentSentence,
+                        start: startTime,
+                        end: endTime
+                    });
+                }
+                currentSentence = "";
+                startTime = w.start;
+            }
+            currentSpeaker = w.speaker;
+            currentSentence += (currentSentence ? " " : "") + w.text; //sentence beginning or not
+            const lastWord = w.text.trim();
+            const lastChar = lastWord.slice(-1);
+            const isAbbreviation = ABBREVIATIONS.has(lastWord);
+
+            //sentence ending
+            if (ENDINGS.includes(lastChar) && !isAbbreviation) {
+                const lastEntry = result[result.length - 1];
+                if (lastEntry && lastEntry.speaker === currentSpeaker) {
+                    lastEntry.sentence += " " + currentSentence;
+                    lastEntry.end = endTime;
+                } else {
+                    result.push({
+                        speaker: currentSpeaker,
+                        sentence: currentSentence,
+                        start: startTime,
+                        end: endTime
+                    });
+                }
+                currentSentence = "";
+                startTime = null;
+                endTime = null;
+                currentSpeaker = null;
+            }
+        }
+
+        // safe last sentence 
+        if (currentSentence) {
+            const lastEntry = result[result.length - 1];
+            if (lastEntry && lastEntry.speaker === currentSpeaker) {
+                lastEntry.sentence += " " + currentSentence;
+                lastEntry.end = endTime;
+            } else {
+                result.push({
+                    speaker: currentSpeaker,
+                    sentence: currentSentence,
+                    start: startTime,
+                    end: endTime
+                });
+            }
+        }
+
+        // Output as Text
+        const output = result.map(r => 
+            `Sprecher ${r.speaker} [${r.start.toFixed(2)} - ${r.end.toFixed(2)}]: ${r.sentence}`
+        );
+
+        // Output on cosole
+        //console.log("\n------------\nMerged Transcription Result:\n", output, "\n------------\n");
+
+        try {
+            const jsonPath = path.join(outputDir, "transcription_result.json");
+            fs.writeFileSync(jsonPath, JSON.stringify(result, null, 2), "utf-8");
+
+            const txtPath = path.join(outputDir, "transcription_result.txt");
+            fs.writeFileSync(txtPath, output.join("\n"), "utf-8");
+
+            console.log(`Summary successfully saved:\n- ${jsonPath}\n- ${txtPath}`);
+        } catch (err) {
+            console.error("Error saving Summary:", err);
+        }
+    }
+}

services/modules/jsonTools/transcriptionSummarizer2.js

@@ -0,0 +1,121 @@
+const fs = require("fs");
+const path = require("path");
+
+// Prepare output directory (always storage/transcriptionSummaries under project root)
+const outputDir = `${__dirname}/../../../storage/transcriptionSummaries`;
+if (!fs.existsSync(outputDir)) {
+    fs.mkdirSync(outputDir, { recursive: true });
+}
+
+//Speaker, Sentence, Start, End
+
+module.exports = {
+    name: "summarize-transcription2", // Unique name for our function that will later be used to get the function from the map via "mapFunctions.get("example").function()"
+    type: "summarizer", // value used to differentiate each module to order them in the UI
+    displayname: "Summarizer",  // The displayname used within the UI
+    async function(args) {
+        let inputJson = args.json;
+
+        //JSON Path
+        if (args.jsonPath) {
+            try {
+                const raw = fs.readFileSync(args.jsonPath, "utf-8");
+                inputJson = JSON.parse(raw);
+            } catch (e) {
+                console.error("Failed to load JSON from file:", e);
+                return { error: "Could not read JSON from file path." };
+            }
+        }
+        // JSON parsen
+        if (typeof inputJson === "string") {
+            try {
+                inputJson = JSON.parse(inputJson);
+            } catch (e) {
+                console.log("Invalid JSON in summarize-transcription");
+                return { error: "Invalid JSON" };
+            }
+        }
+
+        const words = inputJson.words;
+        if (!Array.isArray(words)) {
+            return { error: "No words Array found" };
+        }
+
+        const ENDINGS = [".", "!", "?"]; // '...' auch als Satzende ?
+        const ABBREVIATIONS = new Set(["z.B.", "bzw.", "u.a.", "Dr.", "Mr.", "Mrs.", "Prof.", "etc."]); //TODO weitere Ergaenzen
+
+        const result = [];
+        let currentSentence = "";
+        let currentSpeaker = null;
+        let startTime = null;
+        let endTime = null;
+
+        for (const w of words) {
+            if (!currentSpeaker) currentSpeaker = w.speaker;
+            if (startTime === null) startTime = w.start;
+            endTime = w.end;
+    
+            //speaker changing
+             if (currentSpeaker !== w.speaker && currentSentence) {
+                result.push({
+                    speaker: currentSpeaker,
+                    sentence: currentSentence,
+                    start: startTime,
+                    end: endTime
+                });
+                currentSentence = "";
+                startTime = w.start;
+            }
+            currentSpeaker = w.speaker;
+            currentSentence += (currentSentence ? " " : "") + w.text; //sentence beginning or not
+            const lastWord = w.text.trim();
+            const lastChar = lastWord.slice(-1);
+            const isAbbreviation = ABBREVIATIONS.has(lastWord);
+
+            //sentence ending
+            if (ENDINGS.includes(lastChar) && !isAbbreviation) {
+                result.push({
+                    speaker: currentSpeaker,
+                    sentence: currentSentence,
+                    start: startTime,
+                    end: endTime
+                });
+                currentSentence = "";
+                startTime = null;
+                endTime = null;
+                currentSpeaker = null;
+            }
+        }
+
+        // safe last sentence 
+        if (currentSentence) {
+            result.push({
+                speaker: currentSpeaker,
+                sentence: currentSentence,
+                start: startTime,
+                end: endTime
+            });
+        }
+
+
+        // Output as Text
+        const output = result.map(r => 
+            `Sprecher ${r.speaker} [${r.start.toFixed(2)} - ${r.end.toFixed(2)}]: ${r.sentence}`
+        );
+
+        // Output on cosole
+        //console.log("\n------------\nMerged Transcription Result:\n", output, "\n------------\n");
+
+        try {
+            const jsonPath = path.join(outputDir, "transcription_result.json");
+            fs.writeFileSync(jsonPath, JSON.stringify(result, null, 2), "utf-8");
+
+            const txtPath = path.join(outputDir, "transcription_result.txt");
+            fs.writeFileSync(txtPath, output.join("\n"), "utf-8");
+
+            console.log(`Summary successfully saved:\n- ${jsonPath}\n- ${txtPath}`);
+        } catch (err) {
+            console.error("Error saving Summary:", err);
+        }
+    }
+}

services/modules/transcription-local/whisperLocal.ts

@@ -5,7 +5,7 @@ import { fileURLToPath } from "url"; // To handle __dirname in ES modules
 
 const __filename = fileURLToPath(import.meta.url); // Get current file path
 const __dirname = path.dirname(__filename); // Get current directory path
-const transcriptsDir = path.resolve(__dirname, "../../storage/transcriptions");
+const transcriptsDir = path.resolve(__dirname, "../../../storage/transcriptions");
 
 
 export class whisperLocal { // is called by transcribe.ts
@@ -26,7 +26,6 @@ export class whisperLocal { // is called by transcribe.ts
 
   async transcribe(audioPath: string): Promise<string> { //asyncronous function to transcribe audio
     return new Promise((resolve, reject) => {
-        const transcriptsDir = path.resolve(__dirname, "../../../../storage/transcripts"); //storage directory for transcripts
 
         if (!fs.existsSync(transcriptsDir)) { //if transcripts directory does not exist, create it
             fs.mkdirSync(transcriptsDir, { recursive: true });

services/modules/transcription-remote/assembly.js

@@ -0,0 +1,123 @@
+require('dotenv').config();
+
+const API_KEY = process.env.API_KEY;
+const BASE_URL = 'https://api.assemblyai.com/v2';
+
+//---------------------------------------------------Upload audio---------------------------------------------------
+
+async function uploadAudio(audioPath) {
+  const audioData = fs.readFileSync(audioPath);
+
+  const response = await axios.post(`${BASE_URL}/upload`, audioData, {
+    headers: {
+      authorization: API_KEY,
+      'content-type': 'application/octet-stream'
+    }
+  });
+
+  return response.data.upload_url;
+}
+
+////---------------------------------------------------Extract session id---------------------------------------------------
+
+function getSessionId(inputPath) {
+  try {
+    const parsed = new URL(inputPath);
+    const base = path.basename(parsed.pathname);
+    return base.replace(/\.[^.]+$/, '');
+  } catch {
+    return path.basename(inputPath, path.extname(inputPath));
+  }
+}
+
+//---------------------------------------------------Create transcript---------------------------------------------------
+
+async function createTranscript(audioUrl) {
+  const response = await axios.post(
+    `${BASE_URL}/transcript`,
+    {
+      audio_url: audioUrl,
+      speaker_labels: true,
+      language_detection: true
+    },
+    {
+      headers: {
+        authorization: API_KEY,
+        'content-type': 'application/json'
+      }
+    }
+  );
+
+  return response.data.id;
+}
+
+//---------------------------------------------------Poll transcript---------------------------------------------------
+
+async function pollTranscript(transcriptId) {
+  while (true) {
+    const response = await axios.get(`${BASE_URL}/transcript/${transcriptId}`, {
+      headers: { authorization: API_KEY }
+    });
+
+    const status = response.data.status;
+
+    if (status === 'completed') return response.data;
+    if (status === 'error') throw new Error(`Transcription failed: ${response.data.error}`);
+
+    await new Promise(res => setTimeout(res, 3000));
+  }
+}
+
+//---------------------------------------------------Save transcript---------------------------------------------------
+
+function saveTranscript(transcript, sessionId) {
+  const outputDir = path.join(__dirname, '../../../storage/transcripts');
+
+  if (!fs.existsSync(outputDir)) {
+    fs.mkdirSync(outputDir, { recursive: true });
+  }
+
+  const outputPath = path.join(outputDir, `${sessionId}.json`);
+  fs.writeFileSync(outputPath, JSON.stringify(transcript, null, 2));
+
+  console.log(`Transcript saved: ${outputPath}`);
+}
+
+//---------------------------------------------------Modul---------------------------------------------------
+
+module.exports = {
+  name: 'assembly',
+  type: 'transcription',
+  displayname: 'AssemblyAI',
+
+  async function(audioFileName) {
+    try {
+      // audioFileName ist nur "datei.mp3"
+      const audioPath = path.join(
+        __dirname,
+        '../../../storage/audio',
+        audioFileName
+      );
+
+      let audioUrl;
+
+      if (/^https?:\/\//i.test(audioFileName)) {
+        audioUrl = audioFileName;
+      } else {
+        if (!fs.existsSync(audioPath)) {
+          throw new Error(`Audio file not found: ${audioPath}`);
+        }
+        audioUrl = await uploadAudio(audioPath);
+      }
+
+      const transcriptId = await createTranscript(audioUrl);
+      const transcript = await pollTranscript(transcriptId);
+
+      const sessionId = getSessionId(audioFileName);
+      saveTranscript(transcript, sessionId);
+
+    } catch (error) {
+      console.error('Transcription error:', error.message);
+    }
+  }
+};

services/modules/transcription-remote/assembly.ts

@@ -0,0 +1,133 @@
+import 'dotenv/config';
+import axios from 'axios';
+import fs from 'fs';
+import path from 'path';
+import { fileURLToPath } from 'url';
+
+const __filename = fileURLToPath(import.meta.url);
+const __dirname = path.dirname(__filename);
+
+const API_KEY = process.env.ASSEMBLYAI_API_KEY;
+const BASE_URL = 'https://api.assemblyai.com/v2';
+
+/**
+ * Uploads audio file to AssemblyAI
+ */
+async function uploadAudio(audioPath: string): Promise<string> {
+  const audioData = fs.readFileSync(audioPath);
+  
+  const response = await axios.post<{ upload_url: string }>(`${BASE_URL}/upload`, audioData, {
+    headers: {
+      'authorization': API_KEY,
+      'content-type': 'application/octet-stream'
+    }
+  });
+  
+  return response.data.upload_url;
+}
+
+/**
+ * Extract a session id (basename without extension) from a local path or a URL
+ */
+function getSessionId(inputPath: string): string {
+  try {
+    const parsed = new URL(inputPath);
+    const base = path.basename(parsed.pathname);
+    return base.replace(/\.[^.]+$/, '');
+  } catch (err) {
+    // not a URL, treat as local path
+    return path.basename(inputPath, path.extname(inputPath));
+  }
+}
+
+/**
+ * Creates transcription job with speaker diarization
+ */
+async function createTranscript(audioUrl: string): Promise<string> {
+  const response = await axios.post<{ id: string }>(`${BASE_URL}/transcript`, {
+    audio_url: audioUrl,
+    speaker_labels: true,
+    language_detection: true
+  }, {
+    headers: {
+      'authorization': API_KEY,
+      'content-type': 'application/json'
+    }
+  });
+  
+  return response.data.id;
+}
+
+/**
+ * Polls transcript status until completed
+ */
+async function pollTranscript(transcriptId: string): Promise<any> {
+  while (true) {
+    const response = await axios.get<any>(`${BASE_URL}/transcript/${transcriptId}`, {
+      headers: { 'authorization': API_KEY }
+    });
+    
+    const status = response.data.status;
+    
+    if (status === 'completed') {
+      return response.data;
+    } else if (status === 'error') {
+      throw new Error(`Transcription failed: ${response.data.error}`);
+    }
+    
+    // Wait 3 seconds before next poll
+    await new Promise(resolve => setTimeout(resolve, 3000));
+  }
+}
+
+/**
+ * Saves transcript to storage
+ */
+function saveTranscript(transcript: any, sessionId: string): void {
+  const outputDir = path.join(__dirname, '..', '..', '..', 'storage', 'transcripts');
+  
+  if (!fs.existsSync(outputDir)) {
+    fs.mkdirSync(outputDir, { recursive: true });
+  }
+  
+  const outputPath = path.join(outputDir, `${sessionId}.json`);
+  fs.writeFileSync(outputPath, JSON.stringify(transcript, null, 2));
+  
+  console.log(`✅ Transcript saved: ${outputPath}`);
+}
+
+export default {
+  name: "assembly",
+  type: "transcription",
+  displayname: "AssemblyAI",
+  run: async (audioPath: string) => {
+    try {
+      // Determine if audioPath is an external URL or a local file
+      let audioUrl: string;
+      if (/^https?:\/\//i.test(audioPath)) {
+        console.log('🔗 Using external audio URL...');
+        audioUrl = audioPath;
+      } else {
+        console.log('🔄 Uploading local audio...');
+        if (!fs.existsSync(audioPath)) {
+          throw new Error(`Audio file not found: ${audioPath}`);
+        }
+        audioUrl = await uploadAudio(audioPath);
+      }
+      
+      console.log('🔄 Creating transcript job...');
+      const transcriptId = await createTranscript(audioUrl);
+      
+      console.log('⏳ Waiting for transcription...');
+      const transcript = await pollTranscript(transcriptId);
+      
+  const sessionId = getSessionId(audioPath);
+      saveTranscript(transcript, sessionId);
+      
+      return transcript;
+    } catch (error: any) {
+      console.error('❌ Transcription error:', error.message);
+      throw error;
+    }
+  }
+};

services/modules/transcription/assembly.js

@@ -1,8 +0,0 @@
-module.exports = {
-    name:"assembly", // Unique name for our function that will later be used to get the function from the map via "mapFunctions.get("example").function()"
-    type:"transcription", // value used to differentiate each module to order them in the UI
-    displayname:"Assembly", // The displayname used within the UI
-    async function(parameter){
-        // TODO add code to actually process the audio file
-    }
-}

services/modules/utility/@startup.js

@@ -6,6 +6,13 @@ module.exports = {
         // We are now calling the example function from the example folder
         mapFunctions.get("example").function("Startup")
 
+       let transcript = await mapFunctions.get("assembly").function('../../storage/audio/IMG_2978.wav');
+
+       let summary = await mapFunctions.get("summarize-transcription").function({jsonPath:'/Users/santa/Proj25/video2document/storage/transcripts/IMG_2978.json'});
+
+
+        
+
         // mapFunctions.get("extraction-video-to-audio").function({inputVideoPath:"./a.mp4", outputType:"wav"})
         // mapFunctions.get("extraction-video-to-audio").function({inputVideoPath:"./b.mp4", outputType:"wav"})
         // mapFunctions.get("extraction-video-to-audio").function({inputVideoPath:"./b.mp4", outputType:"flac"})

services/pipeline/jobs/transcribeLatest.ts

@@ -0,0 +1,52 @@
+// services/pipeline/jobs/transcribeLatest.ts
+import path from 'path';
+import fs from 'fs';
+import assembly from '../../modules/transcription/assembly';
+
+/**
+ * Finds the most recently modified .wav file in storage/audio/
+ */
+function getLatestWav(): string {
+  const audioDir = path.join(process.cwd(), 'storage', 'audio');
+  const files = fs.readdirSync(audioDir).filter(f => f.toLowerCase().endsWith('.wav'));
+  if (files.length === 0) throw new Error('⚠️ No .wav file found in storage/audio');
+
+  const newest = files
+    .map(f => ({ f, t: fs.statSync(path.join(audioDir, f)).mtimeMs }))
+    .sort((a, b) => b.t - a.t)[0].f;
+
+  return path.join(audioDir, newest);
+}
+
+/**
+ * Full transcription pipeline according to the defined workflow:
+ * 1. Audio Upload → AssemblyAI
+ * 2. Job Creation (transcript_id)
+ * 3. Polling Status (queued → processing → completed)
+ * 4. Download Transcript JSON
+ * 5. Storage: /transcripts/{session_id}.json
+ */
+async function main() {
+  const audioPath = getLatestWav();
+
+  console.log('1️⃣ Audio Upload → AssemblyAI');
+  console.log('   Source:', audioPath);
+
+  console.log('2️⃣ Job Creation (transcript_id)');
+  console.log('3️⃣ Polling Status (queued → processing → completed)');
+  console.log('4️⃣ Download Transcript JSON');
+  console.log('5️⃣ Storage: /transcripts/{session_id}.json');
+
+  // Execute the transcription process via the AssemblyAI module
+  const result = await assembly.run(audioPath);
+
+  console.log('✅ Transcription completed successfully');
+  console.log('🆔 Transcript ID:', result.id);
+  console.log('📁 Transcript file saved under: storage/transcripts/');
+}
+
+// Entry point
+main().catch((err) => {
+  console.error('❌ Transcription pipeline failed:', err.message || err);
+  process.exit(1);
+});

test/integration/assembly.test.js

@@ -0,0 +1,14 @@
+import 'dotenv/config';
+import assemblyModule from '../../services/modules/transcription-remote/assembly.ts';
+
+// Test: URL passed as argument OR local file ./storage/audio/test.wav
+const audioPath = process.argv[2] || './storage/audio/test.wav';
+
+assemblyModule.run(audioPath)
+  .then(result => {
+    console.log('✅ Success!');
+    console.log('Transcript ID:', result.id);
+  })
+  .catch(error => {
+    console.error('❌ Error:', error?.message || error);
+  });

test/integration/assemblyTest.js

@@ -0,0 +1,18 @@
+require('dotenv').config();
+const path = require('path');
+const assemblyModule = require('../../services/modules/transcription-remote/assembly.js');
+
+// Audio-Datei oder URL aus Kommandozeile, Standard: test.wav
+const audioPath = process.argv[2] || './storage/audio/IMG_2978.wav';
+
+(async () => {
+  try {
+    const transcript = await assemblyModule.run(audioPath);
+
+    console.log('Transcription succesful');
+    console.log('Transcript ID:', transcript?.id);
+    console.log('Speaker labels:', transcript?.utterances?.length || 0);
+  } catch (error) {
+    console.error('Error in Transcription:', error?.message || error);
+  }
+})();

test/integration/transcriptionSummarizerTest.js

@@ -0,0 +1,12 @@
+const fs = require('fs');
+const transSummarizer = require("../../services/modules/jsonTools/transcriptionSummarizer.js");
+
+// JSON-Datei laden
+const inputJson = JSON.parse(fs.readFileSync("./testFile.json", "utf8"));
+
+// Übergabe an den Summarizer
+transSummarizer.function({
+    json: inputJson
+});
+
+

test/test_health.py

@@ -0,0 +1,8 @@
+from fastapi.testclient import TestClient
+from app.main import app
+
+client = TestClient(app)
+
+def test_health():
+    response = client.get("/health")
+    assert response.status_code == 200