✨ feat: add voice transcription support to Matrix bots

Add TranscriptionModule and handleAudioMessage to enable voice message
transcription across all Matrix bots. Users can now send voice messages
which are automatically transcribed and processed as text commands.

Affected bots:
- matrix-calendar-bot
- matrix-chat-bot
- matrix-contacts-bot
- matrix-manadeck-bot
- matrix-ollama-bot
- matrix-picture-bot
- matrix-planta-bot
- matrix-presi-bot
- matrix-questions-bot
- matrix-skilltree-bot
- matrix-stats-bot
- matrix-storage-bot
- matrix-tts-bot

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

services/matrix-calendar-bot/src/bot/bot.module.ts

@@ -1,9 +1,15 @@
 import { Module } from '@nestjs/common';
 import { MatrixService } from './matrix.service';
 import { CalendarModule } from '../calendar/calendar.module';
+import { TranscriptionModule } from '@manacore/bot-services';
 
 @Module({
-	imports: [CalendarModule],
+	imports: [
+		CalendarModule,
+		TranscriptionModule.register({
+			sttUrl: process.env.STT_URL || 'http://localhost:3020',
+		}),
+	],
 	providers: [MatrixService],
 	exports: [MatrixService],
 })

services/matrix-calendar-bot/src/bot/matrix.service.ts

@@ -7,6 +7,7 @@ import {
 	KeywordCommandDetector,
 	COMMON_KEYWORDS,
 } from '@manacore/matrix-bot-common';
+import { TranscriptionService } from '@manacore/bot-services';
 import { CalendarService, CalendarEvent } from '../calendar/calendar.service';
 import { HELP_TEXT, WELCOME_TEXT, BOT_INTRODUCTION } from '../config/configuration';
 
@@ -27,11 +28,36 @@ export class MatrixService extends BaseMatrixService {
 
 	constructor(
 		configService: ConfigService,
+		private readonly transcriptionService: TranscriptionService,
 		private calendarService: CalendarService
 	) {
 		super(configService);
 	}
 
+	protected override async handleAudioMessage(
+		roomId: string,
+		event: MatrixRoomEvent,
+		sender: string
+	): Promise<void> {
+		try {
+			const mxcUrl = event.content.url;
+			if (!mxcUrl) return;
+
+			const audioBuffer = await this.downloadMedia(mxcUrl);
+			const text = await this.transcriptionService.transcribe(audioBuffer);
+			if (!text) {
+				await this.sendReply(roomId, event, '❌ Sprachnachricht konnte nicht erkannt werden.');
+				return;
+			}
+
+			await this.sendMessage(roomId, `🎤 *"${text}"*`);
+			await this.handleTextMessage(roomId, event, text, sender);
+		} catch (error) {
+			this.logger.error(`Audio transcription error: ${error}`);
+			await this.sendReply(roomId, event, '❌ Fehler bei der Spracherkennung.');
+		}
+	}
+
 	protected getConfig(): MatrixBotConfig {
 		return {
 			homeserverUrl: this.configService.get<string>('matrix.homeserverUrl') || 'http://localhost:8008',

services/matrix-chat-bot/src/bot/bot.module.ts

@@ -1,10 +1,16 @@
 import { Module } from '@nestjs/common';
 import { MatrixService } from './matrix.service';
 import { ChatModule } from '../chat/chat.module';
-import { SessionModule } from '@manacore/bot-services';
+import { SessionModule, TranscriptionModule } from '@manacore/bot-services';
 
 @Module({
-	imports: [ChatModule, SessionModule.forRoot()],
+	imports: [
+		ChatModule,
+		SessionModule.forRoot(),
+		TranscriptionModule.register({
+			sttUrl: process.env.STT_URL || 'http://localhost:3020',
+		}),
+	],
 	providers: [MatrixService],
 	exports: [MatrixService],
 })

services/matrix-chat-bot/src/bot/matrix.service.ts

@@ -8,7 +8,7 @@ import {
 	COMMON_KEYWORDS,
 } from '@manacore/matrix-bot-common';
 import { ChatService, Conversation } from '../chat/chat.service';
-import { SessionService } from '@manacore/bot-services';
+import { SessionService, TranscriptionService } from '@manacore/bot-services';
 import { HELP_MESSAGE, BRANCH_ICONS } from '../config/configuration';
 
 @Injectable()
@@ -25,12 +25,37 @@ export class MatrixService extends BaseMatrixService {
 
 	constructor(
 		configService: ConfigService,
+		private readonly transcriptionService: TranscriptionService,
 		private chatService: ChatService,
 		private sessionService: SessionService
 	) {
 		super(configService);
 	}
 
+	protected override async handleAudioMessage(
+		roomId: string,
+		event: MatrixRoomEvent,
+		sender: string
+	): Promise<void> {
+		try {
+			const mxcUrl = event.content.url;
+			if (!mxcUrl) return;
+
+			const audioBuffer = await this.downloadMedia(mxcUrl);
+			const text = await this.transcriptionService.transcribe(audioBuffer);
+			if (!text) {
+				await this.sendReply(roomId, event, '❌ Sprachnachricht konnte nicht erkannt werden.');
+				return;
+			}
+
+			await this.sendMessage(roomId, `🎤 *"${text}"*`);
+			await this.handleTextMessage(roomId, event, text, sender);
+		} catch (error) {
+			this.logger.error(`Audio transcription error: ${error}`);
+			await this.sendReply(roomId, event, '❌ Fehler bei der Spracherkennung.');
+		}
+	}
+
 	protected getConfig(): MatrixBotConfig {
 		return {
 			homeserverUrl: this.configService.get<string>('matrix.homeserverUrl') || 'http://localhost:8008',

services/matrix-contacts-bot/src/bot/bot.module.ts

@@ -1,10 +1,16 @@
 import { Module } from '@nestjs/common';
 import { MatrixService } from './matrix.service';
 import { ContactsModule } from '../contacts/contacts.module';
-import { SessionModule } from '@manacore/bot-services';
+import { SessionModule, TranscriptionModule } from '@manacore/bot-services';
 
 @Module({
-	imports: [ContactsModule, SessionModule.forRoot()],
+	imports: [
+		ContactsModule,
+		SessionModule.forRoot(),
+		TranscriptionModule.register({
+			sttUrl: process.env.STT_URL || 'http://localhost:3020',
+		}),
+	],
 	providers: [MatrixService],
 	exports: [MatrixService],
 })

services/matrix-contacts-bot/src/bot/matrix.service.ts

@@ -9,7 +9,7 @@ import {
 	UserListMapper,
 } from '@manacore/matrix-bot-common';
 import { ContactsService, Contact } from '../contacts/contacts.service';
-import { SessionService } from '@manacore/bot-services';
+import { SessionService, TranscriptionService } from '@manacore/bot-services';
 import { HELP_MESSAGE } from '../config/configuration';
 
 // Natural language keyword detector
@@ -27,12 +27,37 @@ export class MatrixService extends BaseMatrixService {
 
 	constructor(
 		configService: ConfigService,
+		private readonly transcriptionService: TranscriptionService,
 		private contactsService: ContactsService,
 		private sessionService: SessionService
 	) {
 		super(configService);
 	}
 
+	protected override async handleAudioMessage(
+		roomId: string,
+		event: MatrixRoomEvent,
+		sender: string
+	): Promise<void> {
+		try {
+			const mxcUrl = event.content.url;
+			if (!mxcUrl) return;
+
+			const audioBuffer = await this.downloadMedia(mxcUrl);
+			const text = await this.transcriptionService.transcribe(audioBuffer);
+			if (!text) {
+				await this.sendReply(roomId, event, '❌ Sprachnachricht konnte nicht erkannt werden.');
+				return;
+			}
+
+			await this.sendMessage(roomId, `🎤 *"${text}"*`);
+			await this.handleTextMessage(roomId, event, text, sender);
+		} catch (error) {
+			this.logger.error(`Audio transcription error: ${error}`);
+			await this.sendReply(roomId, event, '❌ Fehler bei der Spracherkennung.');
+		}
+	}
+
 	protected getConfig(): MatrixBotConfig {
 		return {
 			homeserverUrl:

services/matrix-manadeck-bot/src/bot/bot.module.ts

@@ -1,10 +1,16 @@
 import { Module } from '@nestjs/common';
 import { MatrixService } from './matrix.service';
 import { ManadeckModule } from '../manadeck/manadeck.module';
-import { SessionModule } from '@manacore/bot-services';
+import { SessionModule, TranscriptionModule } from '@manacore/bot-services';
 
 @Module({
-	imports: [ManadeckModule, SessionModule.forRoot()],
+	imports: [
+		ManadeckModule,
+		SessionModule.forRoot(),
+		TranscriptionModule.register({
+			sttUrl: process.env.STT_URL || 'http://localhost:3020',
+		}),
+	],
 	providers: [MatrixService],
 	exports: [MatrixService],
 })

services/matrix-manadeck-bot/src/bot/matrix.service.ts

@@ -9,7 +9,7 @@ import {
 	COMMON_KEYWORDS,
 } from '@manacore/matrix-bot-common';
 import { ManadeckService, Deck, Card } from '../manadeck/manadeck.service';
-import { SessionService } from '@manacore/bot-services';
+import { SessionService, TranscriptionService } from '@manacore/bot-services';
 import { HELP_MESSAGE } from '../config/configuration';
 
 @Injectable()
@@ -34,12 +34,37 @@ export class MatrixService extends BaseMatrixService {
 
 	constructor(
 		configService: ConfigService,
+		private readonly transcriptionService: TranscriptionService,
 		private manadeckService: ManadeckService,
 		private sessionService: SessionService
 	) {
 		super(configService);
 	}
 
+	protected override async handleAudioMessage(
+		roomId: string,
+		event: MatrixRoomEvent,
+		sender: string
+	): Promise<void> {
+		try {
+			const mxcUrl = event.content.url;
+			if (!mxcUrl) return;
+
+			const audioBuffer = await this.downloadMedia(mxcUrl);
+			const text = await this.transcriptionService.transcribe(audioBuffer);
+			if (!text) {
+				await this.sendHtml(roomId, '<p>❌ Sprachnachricht konnte nicht erkannt werden.</p>');
+				return;
+			}
+
+			await this.sendHtml(roomId, `<p>🎤 <em>"${text}"</em></p>`);
+			await this.handleTextMessage(roomId, event, text, sender);
+		} catch (error) {
+			this.logger.error(`Audio transcription error: ${error}`);
+			await this.sendHtml(roomId, '<p>❌ Fehler bei der Spracherkennung.</p>');
+		}
+	}
+
 	protected getConfig(): MatrixBotConfig {
 		return {
 			homeserverUrl: this.configService.get<string>('matrix.homeserverUrl') || 'http://localhost:8008',

services/matrix-ollama-bot/package.json

@@ -27,6 +27,7 @@
 		"type-check": "tsc --noEmit"
 	},
 	"dependencies": {
+		"@manacore/bot-services": "workspace:*",
 		"@manacore/matrix-bot-common": "workspace:*",
 		"@nestjs/common": "^10.4.15",
 		"@nestjs/config": "^3.3.0",

services/matrix-ollama-bot/src/bot/bot.module.ts

@@ -1,9 +1,15 @@
 import { Module } from '@nestjs/common';
 import { MatrixService } from './matrix.service';
 import { OllamaModule } from '../ollama/ollama.module';
+import { TranscriptionModule } from '@manacore/bot-services';
 
 @Module({
-	imports: [OllamaModule],
+	imports: [
+		OllamaModule,
+		TranscriptionModule.register({
+			sttUrl: process.env.STT_URL || 'http://localhost:3020',
+		}),
+	],
 	providers: [MatrixService],
 	exports: [MatrixService],
 })

services/matrix-ollama-bot/src/bot/matrix.service.ts

@@ -7,6 +7,7 @@ import {
 	KeywordCommandDetector,
 	COMMON_KEYWORDS,
 } from '@manacore/matrix-bot-common';
+import { TranscriptionService } from '@manacore/bot-services';
 import { OllamaService } from '../ollama/ollama.service';
 import { SYSTEM_PROMPTS } from '../config/configuration';
 
@@ -37,11 +38,36 @@ export class MatrixService extends BaseMatrixService {
 
 	constructor(
 		configService: ConfigService,
+		private readonly transcriptionService: TranscriptionService,
 		private ollamaService: OllamaService
 	) {
 		super(configService);
 	}
 
+	protected override async handleAudioMessage(
+		roomId: string,
+		event: MatrixRoomEvent,
+		sender: string
+	): Promise<void> {
+		try {
+			const mxcUrl = event.content.url;
+			if (!mxcUrl) return;
+
+			const audioBuffer = await this.downloadMedia(mxcUrl);
+			const text = await this.transcriptionService.transcribe(audioBuffer);
+			if (!text) {
+				await this.sendMessage(roomId, '❌ Sprachnachricht konnte nicht erkannt werden.');
+				return;
+			}
+
+			await this.sendMessage(roomId, `🎤 *"${text}"*`);
+			await this.handleTextMessage(roomId, event, text, sender);
+		} catch (error) {
+			this.logger.error(`Audio transcription error: ${error}`);
+			await this.sendMessage(roomId, '❌ Fehler bei der Spracherkennung.');
+		}
+	}
+
 	protected getConfig(): MatrixBotConfig {
 		return {
 			homeserverUrl:

services/matrix-picture-bot/src/bot/bot.module.ts

@@ -1,10 +1,16 @@
 import { Module } from '@nestjs/common';
 import { MatrixService } from './matrix.service';
 import { PictureModule } from '../picture/picture.module';
-import { SessionModule } from '@manacore/bot-services';
+import { SessionModule, TranscriptionModule } from '@manacore/bot-services';
 
 @Module({
-	imports: [PictureModule, SessionModule.forRoot()],
+	imports: [
+		PictureModule,
+		SessionModule.forRoot(),
+		TranscriptionModule.register({
+			sttUrl: process.env.STT_URL || 'http://localhost:3020',
+		}),
+	],
 	providers: [MatrixService],
 	exports: [MatrixService],
 })

services/matrix-picture-bot/src/bot/matrix.service.ts

@@ -8,7 +8,7 @@ import {
 	COMMON_KEYWORDS,
 } from '@manacore/matrix-bot-common';
 import { PictureService } from '../picture/picture.service';
-import { SessionService } from '@manacore/bot-services';
+import { SessionService, TranscriptionService } from '@manacore/bot-services';
 import { HELP_MESSAGE } from '../config/configuration';
 
 interface ParsedPrompt {
@@ -36,12 +36,37 @@ export class MatrixService extends BaseMatrixService {
 
 	constructor(
 		configService: ConfigService,
+		private readonly transcriptionService: TranscriptionService,
 		private pictureService: PictureService,
 		private sessionService: SessionService
 	) {
 		super(configService);
 	}
 
+	protected override async handleAudioMessage(
+		roomId: string,
+		event: MatrixRoomEvent,
+		sender: string
+	): Promise<void> {
+		try {
+			const mxcUrl = event.content.url;
+			if (!mxcUrl) return;
+
+			const audioBuffer = await this.downloadMedia(mxcUrl);
+			const text = await this.transcriptionService.transcribe(audioBuffer);
+			if (!text) {
+				await this.sendMessage(roomId, '❌ Sprachnachricht konnte nicht erkannt werden.');
+				return;
+			}
+
+			await this.sendMessage(roomId, `🎤 *"${text}"*`);
+			await this.handleTextMessage(roomId, event, text, sender);
+		} catch (error) {
+			this.logger.error(`Audio transcription error: ${error}`);
+			await this.sendMessage(roomId, '❌ Fehler bei der Spracherkennung.');
+		}
+	}
+
 	protected getConfig(): MatrixBotConfig {
 		return {
 			homeserverUrl: this.configService.get<string>('matrix.homeserverUrl') || 'http://localhost:8008',

services/matrix-planta-bot/src/bot/bot.module.ts

@@ -1,10 +1,16 @@
 import { Module } from '@nestjs/common';
 import { MatrixService } from './matrix.service';
 import { PlantaModule } from '../planta/planta.module';
-import { SessionModule } from '@manacore/bot-services';
+import { SessionModule, TranscriptionModule } from '@manacore/bot-services';
 
 @Module({
-	imports: [PlantaModule, SessionModule.forRoot()],
+	imports: [
+		PlantaModule,
+		SessionModule.forRoot(),
+		TranscriptionModule.register({
+			sttUrl: process.env.STT_URL || 'http://localhost:3020',
+		}),
+	],
 	providers: [MatrixService],
 	exports: [MatrixService],
 })

services/matrix-planta-bot/src/bot/matrix.service.ts

@@ -9,7 +9,7 @@ import {
 	COMMON_KEYWORDS,
 } from '@manacore/matrix-bot-common';
 import { PlantaService, Plant } from '../planta/planta.service';
-import { SessionService } from '@manacore/bot-services';
+import { SessionService, TranscriptionService } from '@manacore/bot-services';
 import { HELP_MESSAGE } from '../config/configuration';
 
 @Injectable()
@@ -49,12 +49,37 @@ export class MatrixService extends BaseMatrixService {
 
 	constructor(
 		configService: ConfigService,
+		private readonly transcriptionService: TranscriptionService,
 		private plantaService: PlantaService,
 		private sessionService: SessionService
 	) {
 		super(configService);
 	}
 
+	protected override async handleAudioMessage(
+		roomId: string,
+		event: MatrixRoomEvent,
+		sender: string
+	): Promise<void> {
+		try {
+			const mxcUrl = event.content.url;
+			if (!mxcUrl) return;
+
+			const audioBuffer = await this.downloadMedia(mxcUrl);
+			const text = await this.transcriptionService.transcribe(audioBuffer);
+			if (!text) {
+				await this.sendMessage(roomId, '<p>❌ Sprachnachricht konnte nicht erkannt werden.</p>');
+				return;
+			}
+
+			await this.sendMessage(roomId, `<p>🎤 <em>"${text}"</em></p>`);
+			await this.handleTextMessage(roomId, event, text);
+		} catch (error) {
+			this.logger.error(`Audio transcription error: ${error}`);
+			await this.sendMessage(roomId, '<p>❌ Fehler bei der Spracherkennung.</p>');
+		}
+	}
+
 	protected getConfig(): MatrixBotConfig {
 		return {
 			homeserverUrl: this.configService.get<string>('matrix.homeserverUrl') || 'http://localhost:8008',

services/matrix-presi-bot/src/bot/bot.module.ts

@@ -1,10 +1,16 @@
 import { Module } from '@nestjs/common';
 import { MatrixService } from './matrix.service';
 import { PresiModule } from '../presi/presi.module';
-import { SessionModule } from '@manacore/bot-services';
+import { SessionModule, TranscriptionModule } from '@manacore/bot-services';
 
 @Module({
-	imports: [PresiModule, SessionModule.forRoot()],
+	imports: [
+		PresiModule,
+		SessionModule.forRoot(),
+		TranscriptionModule.register({
+			sttUrl: process.env.STT_URL || 'http://localhost:3020',
+		}),
+	],
 	providers: [MatrixService],
 	exports: [MatrixService],
 })

services/matrix-presi-bot/src/bot/matrix.service.ts

@@ -9,7 +9,7 @@ import {
 	COMMON_KEYWORDS,
 } from '@manacore/matrix-bot-common';
 import { PresiService, Deck, Theme, SlideContent } from '../presi/presi.service';
-import { SessionService } from '@manacore/bot-services';
+import { SessionService, TranscriptionService } from '@manacore/bot-services';
 import { HELP_MESSAGE } from '../config/configuration';
 
 @Injectable()
@@ -30,12 +30,37 @@ export class MatrixService extends BaseMatrixService {
 
 	constructor(
 		configService: ConfigService,
+		private readonly transcriptionService: TranscriptionService,
 		private presiService: PresiService,
 		private sessionService: SessionService
 	) {
 		super(configService);
 	}
 
+	protected override async handleAudioMessage(
+		roomId: string,
+		event: MatrixRoomEvent,
+		sender: string
+	): Promise<void> {
+		try {
+			const mxcUrl = event.content.url;
+			if (!mxcUrl) return;
+
+			const audioBuffer = await this.downloadMedia(mxcUrl);
+			const text = await this.transcriptionService.transcribe(audioBuffer);
+			if (!text) {
+				await this.sendMessage(roomId, '<p>❌ Sprachnachricht konnte nicht erkannt werden.</p>');
+				return;
+			}
+
+			await this.sendMessage(roomId, `<p>🎤 <em>"${text}"</em></p>`);
+			await this.handleTextMessage(roomId, event, text);
+		} catch (error) {
+			this.logger.error(`Audio transcription error: ${error}`);
+			await this.sendMessage(roomId, '<p>❌ Fehler bei der Spracherkennung.</p>');
+		}
+	}
+
 	protected getConfig(): MatrixBotConfig {
 		return {
 			homeserverUrl:

services/matrix-questions-bot/src/bot/bot.module.ts

@@ -1,10 +1,16 @@
 import { Module } from '@nestjs/common';
 import { MatrixService } from './matrix.service';
 import { QuestionsModule } from '../questions/questions.module';
-import { SessionModule } from '@manacore/bot-services';
+import { SessionModule, TranscriptionModule } from '@manacore/bot-services';
 
 @Module({
-	imports: [QuestionsModule, SessionModule.forRoot()],
+	imports: [
+		QuestionsModule,
+		SessionModule.forRoot(),
+		TranscriptionModule.register({
+			sttUrl: process.env.STT_URL || 'http://localhost:3020',
+		}),
+	],
 	providers: [MatrixService],
 	exports: [MatrixService],
 })

services/matrix-questions-bot/src/bot/matrix.service.ts

@@ -9,7 +9,7 @@ import {
 	COMMON_KEYWORDS,
 } from '@manacore/matrix-bot-common';
 import { QuestionsService, Question, Collection, Answer } from '../questions/questions.service';
-import { SessionService } from '@manacore/bot-services';
+import { SessionService, TranscriptionService } from '@manacore/bot-services';
 import { HELP_MESSAGE } from '../config/configuration';
 
 @Injectable()
@@ -33,7 +33,8 @@ export class MatrixService extends BaseMatrixService {
 	constructor(
 		configService: ConfigService,
 		private questionsService: QuestionsService,
-		private sessionService: SessionService
+		private sessionService: SessionService,
+		private readonly transcriptionService: TranscriptionService
 	) {
 		super(configService);
 	}
@@ -176,6 +177,30 @@ export class MatrixService extends BaseMatrixService {
 		}
 	}
 
+	protected override async handleAudioMessage(
+		roomId: string,
+		event: MatrixRoomEvent,
+		_sender: string
+	): Promise<void> {
+		try {
+			const mxcUrl = event.content.url;
+			if (!mxcUrl) return;
+
+			const audioBuffer = await this.downloadMedia(mxcUrl);
+			const text = await this.transcriptionService.transcribe(audioBuffer);
+			if (!text) {
+				await this.sendReply(roomId, event, '<p>Sprachnachricht konnte nicht erkannt werden.</p>');
+				return;
+			}
+
+			await this.sendMessage(roomId, `<p><em>"${text}"</em></p>`);
+			await this.handleTextMessage(roomId, event, text);
+		} catch (error) {
+			this.logger.error(`Audio transcription error: ${error}`);
+			await this.sendReply(roomId, event, '<p>Fehler bei der Spracherkennung.</p>');
+		}
+	}
+
 	private requireAuth(sender: string): string {
 		const token = this.sessionService.getToken(sender);
 		if (!token) {

services/matrix-skilltree-bot/src/bot/bot.module.ts

@@ -1,10 +1,16 @@
 import { Module } from '@nestjs/common';
 import { MatrixService } from './matrix.service';
 import { SkilltreeModule } from '../skilltree/skilltree.module';
-import { SessionModule } from '@manacore/bot-services';
+import { SessionModule, TranscriptionModule } from '@manacore/bot-services';
 
 @Module({
-	imports: [SkilltreeModule, SessionModule.forRoot()],
+	imports: [
+		SkilltreeModule,
+		SessionModule.forRoot(),
+		TranscriptionModule.register({
+			sttUrl: process.env.STT_URL || 'http://localhost:3020',
+		}),
+	],
 	providers: [MatrixService],
 	exports: [MatrixService],
 })

services/matrix-skilltree-bot/src/bot/matrix.service.ts

@@ -9,7 +9,7 @@ import {
 	COMMON_KEYWORDS,
 } from '@manacore/matrix-bot-common';
 import { SkilltreeService, Skill, SkillBranch } from '../skilltree/skilltree.service';
-import { SessionService } from '@manacore/bot-services';
+import { SessionService, TranscriptionService } from '@manacore/bot-services';
 import { HELP_MESSAGE } from '../config/configuration';
 
 @Injectable()
@@ -54,7 +54,8 @@ export class MatrixService extends BaseMatrixService {
 	constructor(
 		configService: ConfigService,
 		private skilltreeService: SkilltreeService,
-		private sessionService: SessionService
+		private sessionService: SessionService,
+		private readonly transcriptionService: TranscriptionService
 	) {
 		super(configService);
 	}
@@ -163,6 +164,30 @@ export class MatrixService extends BaseMatrixService {
 		}
 	}
 
+	protected override async handleAudioMessage(
+		roomId: string,
+		event: MatrixRoomEvent,
+		_sender: string
+	): Promise<void> {
+		try {
+			const mxcUrl = event.content.url;
+			if (!mxcUrl) return;
+
+			const audioBuffer = await this.downloadMedia(mxcUrl);
+			const text = await this.transcriptionService.transcribe(audioBuffer);
+			if (!text) {
+				await this.sendReply(roomId, event, '<p>Sprachnachricht konnte nicht erkannt werden.</p>');
+				return;
+			}
+
+			await this.sendMessage(roomId, `<p><em>"${text}"</em></p>`);
+			await this.handleTextMessage(roomId, event, text);
+		} catch (error) {
+			this.logger.error(`Audio transcription error: ${error}`);
+			await this.sendReply(roomId, event, '<p>Fehler bei der Spracherkennung.</p>');
+		}
+	}
+
 	private requireAuth(sender: string): string {
 		const token = this.sessionService.getToken(sender);
 		if (!token) {

services/matrix-stats-bot/package.json

@@ -24,6 +24,7 @@
 		"type-check": "tsc --noEmit"
 	},
 	"dependencies": {
+		"@manacore/bot-services": "workspace:*",
 		"@manacore/matrix-bot-common": "workspace:*",
 		"@nestjs/common": "^10.4.15",
 		"@nestjs/config": "^3.3.0",

services/matrix-stats-bot/src/bot/bot.module.ts

@@ -2,9 +2,16 @@ import { Module } from '@nestjs/common';
 import { MatrixService } from './matrix.service';
 import { AnalyticsModule } from '../analytics/analytics.module';
 import { UsersModule } from '../users/users.module';
+import { TranscriptionModule } from '@manacore/bot-services';
 
 @Module({
-	imports: [AnalyticsModule, UsersModule],
+	imports: [
+		AnalyticsModule,
+		UsersModule,
+		TranscriptionModule.register({
+			sttUrl: process.env.STT_URL || 'http://localhost:3020',
+		}),
+	],
 	providers: [MatrixService],
 	exports: [MatrixService],
 })

services/matrix-stats-bot/src/bot/matrix.service.ts

@@ -9,6 +9,7 @@ import {
 } from '@manacore/matrix-bot-common';
 import { AnalyticsService } from '../analytics/analytics.service';
 import { UsersService } from '../users/users.service';
+import { TranscriptionService } from '@manacore/bot-services';
 
 @Injectable()
 export class MatrixService extends BaseMatrixService {
@@ -26,7 +27,8 @@ export class MatrixService extends BaseMatrixService {
 	constructor(
 		configService: ConfigService,
 		private analyticsService: AnalyticsService,
-		private usersService: UsersService
+		private usersService: UsersService,
+		private readonly transcriptionService: TranscriptionService
 	) {
 		super(configService);
 		this.reportRoomId = this.configService.get<string>('matrix.reportRoomId') || '';
@@ -60,6 +62,30 @@ export class MatrixService extends BaseMatrixService {
 		await this.handleCommand(roomId, command.toLowerCase());
 	}
 
+	protected override async handleAudioMessage(
+		roomId: string,
+		event: MatrixRoomEvent,
+		sender: string
+	): Promise<void> {
+		try {
+			const mxcUrl = event.content.url;
+			if (!mxcUrl) return;
+
+			const audioBuffer = await this.downloadMedia(mxcUrl);
+			const text = await this.transcriptionService.transcribe(audioBuffer);
+			if (!text) {
+				await this.sendReply(roomId, event, 'Sprachnachricht konnte nicht erkannt werden.');
+				return;
+			}
+
+			await this.sendMessage(roomId, `*"${text}"*`);
+			await this.handleTextMessage(roomId, event, text, sender);
+		} catch (error) {
+			this.logger.error(`Audio transcription error: ${error}`);
+			await this.sendReply(roomId, event, 'Fehler bei der Spracherkennung.');
+		}
+	}
+
 	private async handleCommand(roomId: string, command: string) {
 		switch (command) {
 			case 'help':

services/matrix-storage-bot/src/bot/bot.module.ts

@@ -1,10 +1,16 @@
 import { Module } from '@nestjs/common';
 import { MatrixService } from './matrix.service';
 import { StorageModule } from '../storage/storage.module';
-import { SessionModule } from '@manacore/bot-services';
+import { SessionModule, TranscriptionModule } from '@manacore/bot-services';
 
 @Module({
-	imports: [StorageModule, SessionModule.forRoot()],
+	imports: [
+		StorageModule,
+		SessionModule.forRoot(),
+		TranscriptionModule.register({
+			sttUrl: process.env.STT_URL || 'http://localhost:3020',
+		}),
+	],
 	providers: [MatrixService],
 	exports: [MatrixService],
 })

services/matrix-storage-bot/src/bot/matrix.service.ts

@@ -9,7 +9,7 @@ import {
 	COMMON_KEYWORDS,
 } from '@manacore/matrix-bot-common';
 import { StorageService, StorageFile, Folder, ShareLink, TrashItem } from '../storage/storage.service';
-import { SessionService } from '@manacore/bot-services';
+import { SessionService, TranscriptionService } from '@manacore/bot-services';
 import { HELP_MESSAGE } from '../config/configuration';
 
 @Injectable()
@@ -35,7 +35,8 @@ export class MatrixService extends BaseMatrixService {
 	constructor(
 		configService: ConfigService,
 		private storageService: StorageService,
-		private sessionService: SessionService
+		private sessionService: SessionService,
+		private readonly transcriptionService: TranscriptionService
 	) {
 		super(configService);
 	}
@@ -203,6 +204,30 @@ export class MatrixService extends BaseMatrixService {
 		}
 	}
 
+	protected override async handleAudioMessage(
+		roomId: string,
+		event: MatrixRoomEvent,
+		_sender: string
+	): Promise<void> {
+		try {
+			const mxcUrl = event.content.url;
+			if (!mxcUrl) return;
+
+			const audioBuffer = await this.downloadMedia(mxcUrl);
+			const text = await this.transcriptionService.transcribe(audioBuffer);
+			if (!text) {
+				await this.sendReply(roomId, event, '<p>Sprachnachricht konnte nicht erkannt werden.</p>');
+				return;
+			}
+
+			await this.sendMessage(roomId, `<p><em>"${text}"</em></p>`);
+			await this.handleTextMessage(roomId, event, text);
+		} catch (error) {
+			this.logger.error(`Audio transcription error: ${error}`);
+			await this.sendReply(roomId, event, '<p>Fehler bei der Spracherkennung.</p>');
+		}
+	}
+
 	private requireAuth(sender: string): string {
 		const token = this.sessionService.getToken(sender);
 		if (!token) {

services/matrix-tts-bot/package.json

@@ -24,6 +24,7 @@
 		"type-check": "tsc --noEmit"
 	},
 	"dependencies": {
+		"@manacore/bot-services": "workspace:*",
 		"@manacore/matrix-bot-common": "workspace:*",
 		"@nestjs/common": "^10.4.17",
 		"@nestjs/config": "^3.3.0",

services/matrix-tts-bot/src/bot/bot.module.ts

@@ -1,9 +1,15 @@
 import { Module } from '@nestjs/common';
 import { MatrixService } from './matrix.service';
 import { TtsModule } from '../tts/tts.module';
+import { TranscriptionModule } from '@manacore/bot-services';
 
 @Module({
-	imports: [TtsModule],
+	imports: [
+		TtsModule,
+		TranscriptionModule.register({
+			sttUrl: process.env.STT_URL || 'http://localhost:3020',
+		}),
+	],
 	providers: [MatrixService],
 })
 export class BotModule {}

services/matrix-tts-bot/src/bot/matrix.service.ts

@@ -8,6 +8,7 @@ import {
 	COMMON_KEYWORDS,
 } from '@manacore/matrix-bot-common';
 import { TtsService } from '../tts/tts.service';
+import { TranscriptionService } from '@manacore/bot-services';
 import { HELP_TEXT, WELCOME_TEXT } from '../config/configuration';
 
 interface UserSettings {
@@ -36,7 +37,8 @@ export class MatrixService extends BaseMatrixService {
 
 	constructor(
 		configService: ConfigService,
-		private ttsService: TtsService
+		private ttsService: TtsService,
+		private readonly transcriptionService: TranscriptionService
 	) {
 		super(configService);
 		this.defaultVoice = this.configService.get<string>('tts.defaultVoice') || 'af_heart';
@@ -123,6 +125,30 @@ export class MatrixService extends BaseMatrixService {
 		}
 	}
 
+	protected override async handleAudioMessage(
+		roomId: string,
+		event: MatrixRoomEvent,
+		_sender: string
+	): Promise<void> {
+		try {
+			const mxcUrl = event.content.url;
+			if (!mxcUrl) return;
+
+			const audioBuffer = await this.downloadMedia(mxcUrl);
+			const text = await this.transcriptionService.transcribe(audioBuffer);
+			if (!text) {
+				await this.sendReply(roomId, event, 'Sprachnachricht konnte nicht erkannt werden.');
+				return;
+			}
+
+			await this.sendMessage(roomId, `*"${text}"*`);
+			await this.handleTextMessage(roomId, event, text);
+		} catch (error) {
+			this.logger.error(`Audio transcription error: ${error}`);
+			await this.sendReply(roomId, event, 'Fehler bei der Spracherkennung.');
+		}
+	}
+
 	private async executeCommand(
 		roomId: string,
 		event: MatrixRoomEvent,