Telegram Voice Translator to Slack

What This Agent Does

This intelligent workflow transforms Telegram voice messages into multilingual Slack notifications with automatic transcription, language detection, and translation capabilities. When a voice message arrives in Telegram, the agent automatically transcribes the audio, detects the original language, translates it to your preferred language, and delivers a beautifully formatted message to your Slack workspace.

Key benefits and time savings:

• Eliminate manual transcription – Voice messages are instantly converted to text, saving hours of manual work
• Break language barriers – Automatic language detection and translation enable seamless global team communication
• Centralized notifications – All processed messages flow directly to Slack, keeping your team informed in one place
• Hands-free workflow – No manual intervention required; the entire process runs automatically from voice to Slack delivery

Target use cases:

• International teams receiving voice updates from multilingual team members
• Customer support teams processing voice feedback and complaints
• Remote teams capturing quick voice notes and sharing them across channels
• Content creators transcribing and translating voice memos for distribution

Who is it for

This workflow is ideal for distributed teams, multilingual organizations, and anyone who receives voice messages across different platforms. It's particularly valuable for:

• Customer success teams managing global clients with diverse language preferences
• Content teams needing quick transcription and translation of voice notes
• Remote-first companies where team members communicate across time zones and languages
• Support departments processing voice-based customer inquiries
• Project managers capturing voice updates and sharing them with stakeholders

No coding experience is required—TaskAGI handles all the complexity while you enjoy the benefits.

Required Integrations

Telegram

Why it's needed: Telegram serves as your voice message source. This integration captures incoming voice messages from your Telegram account or bot, triggering the entire workflow automatically.

Setup steps:

1. Create a Telegram Bot (if you don't have one)

   • Open Telegram and search for @BotFather
   • Send the command /newbot
   • Follow the prompts to name your bot and receive your Bot Token
   • Save this token securely—you'll need it for TaskAGI

2. Obtain Your Chat ID

   • Send a message to your new bot
   • Visit https://api.telegram.org/bot<YOUR_BOT_TOKEN>/getUpdates in your browser
   • Look for the "chat" object and note your Chat ID

3. Connect to TaskAGI

   • Navigate to Integrations → Telegram
   • Click Add New Connection
   • Paste your Bot Token in the authentication field
   • Name this connection (e.g., "My Telegram Bot")
   • Click Save and Test

4. Configure the Telegram Trigger Node

   • In your workflow, select the Telegram Trigger node
   • Choose your Telegram connection from the dropdown
   • Set the Chat ID to your saved Chat ID
   • Enable Webhook mode for real-time message capture
   • Save the configuration

Pro tip: Test your Telegram connection by sending a simple message to your bot before proceeding.

OpenAI

Why it's needed: OpenAI powers two critical functions: transcribing voice messages to text and translating content across languages using GPT-4o-mini.

Setup steps:

1. Create an OpenAI Account

   • Visit https://platform.openai.com/signup
   • Complete registration and verify your email
   • Add a payment method to your account

2. Generate Your API Key

   • Log in to your OpenAI account
   • Navigate to API Keys (in the left sidebar under your profile)
   • Click Create new secret key
   • Copy the key immediately—you won't see it again
   • Store it securely (consider using a password manager)

3. Connect to TaskAGI

   • Go to Integrations → OpenAI
   • Click Add New Connection
   • Paste your API key in the API Key field
   • Name this connection (e.g., "OpenAI Production")
   • Click Save and Test

4. Configure OpenAI Nodes

   • Transcribe Voice node: Select your OpenAI connection and set the model to whisper-1
   • Translate Text node: Select your OpenAI connection and set the model to gpt-4o-mini
   • Both nodes will automatically use your API key for authentication

Budget consideration: Monitor your OpenAI usage in the dashboard to manage costs. Transcription and translation are relatively inexpensive but scale with message volume.

Slack

Why it's needed: Slack is your final destination, where processed messages are delivered to your team in a formatted, actionable way.

Setup steps:

1. Create a Slack App

   • Visit https://api.slack.com/apps
   • Click Create New App → From scratch
   • Name your app (e.g., "TaskAGI Voice Processor")
   • Select your workspace
   • Click Create App

2. Enable Permissions

   • In your app settings, navigate to OAuth & Permissions
   • Under Scopes → Bot Token Scopes, add:
     • chat:write (to send messages)
     • files:read (to access file information)
   • Click Save Scopes

3. Install Your App

   • At the top of the OAuth & Permissions page, click Install to Workspace
   • Authorize the app when prompted
   • Copy your Bot User OAuth Token (starts with xoxb-)

4. Connect to TaskAGI

   • Go to Integrations → Slack
   • Click Add New Connection
   • Paste your Bot User OAuth Token
   • Name this connection (e.g., "Slack Workspace")
   • Click Save and Test

5. Configure the Send to Slack Node

   • Select your Slack connection
   • Choose your target Channel (e.g., #voice-messages or #notifications)
   • The message format will be populated automatically by the Build Slack Message node

Configuration Steps

Node-by-Node Configuration Guidance

Telegram Trigger Node (Node 4)

• Purpose: Captures incoming voice messages from Telegram
• Configuration: Select your Telegram connection and Chat ID
• Output: Raw message data including voice file information

Is Voice Message? (Node 5)

• Purpose: Determines whether the incoming message is a voice message
• Condition: Check if message.voice exists
• True path: Routes to voice processing (Nodes 6-12)
• False path: Routes to no-operation (Node 13)

Extract Voice Info (Node 6)

• Purpose: Extracts relevant metadata from the voice message
• Extract these fields:
  • file_id – Unique identifier for the voice file
  • duration – Length of the voice message in seconds
  • mime_type – Audio format (typically audio/ogg)
• Output: Structured data for the next node

Get Telegram File (Node 7)

• Purpose: Retrieves the actual voice file from Telegram servers
• Configuration: Use the file_id from Node 6
• Output: Direct URL to the audio file

Transcribe Voice (Node 8)

• Purpose: Converts audio to text using OpenAI's Whisper model
• Configuration:
  • Model: whisper-1
  • Input: Audio URL from Node 7
  • Language: Leave blank for auto-detection
• Output: Transcribed text

Detect Language (Node 9)

• Purpose: Identifies the language of the transcribed text
• Configuration: Use a custom function to analyze text and return language code
• Example output: "es" for Spanish, "fr" for French, "en" for English
• Output: Language code for translation

Translate Text (Node 10)

• Purpose: Translates transcribed text to your target language
• Configuration:
  • Model: gpt-4o-mini
  • Prompt: "You are a translation engine. Translate the following text from [detected_language] to English: [transcribed_text]"
  • Replace [detected_language] and [transcribed_text] with actual values
• Output: Translated message text

Build Slack Message (Node 11)

• Purpose: Formats all data into a professional Slack message
• Include these elements:
  • Original transcription
  • Detected language
  • Translated text
  • Timestamp and duration
  • Source indicator (Telegram)
• Output: Formatted message object ready for Slack

Send to Slack (Node 12)

• Purpose: Delivers the formatted message to your Slack channel
• Configuration:
  • Connection: Your Slack connection
  • Channel: Target channel (e.g., #voice-messages)
  • Message: Output from Node 11
• Output: Confirmation of successful delivery

Not a Voice Message (Node 13)

• Purpose: Handles non-voice messages gracefully
• Configuration: No operation—message is ignored
• Output: Workflow completes without error

Testing Your Agent

Step 1: Run Test Execution

1. Send a test voice message to your Telegram bot
2. Click the Test button in TaskAGI's workflow editor
3. Select your test message from the trigger options
4. Click Run Test

Step 2: Verify Each Step

• Node 5 (Is Voice Message?): Confirm the condition evaluates to true
• Node 7 (Get Telegram File): Verify the file URL is accessible
• Node 8 (Transcribe Voice): Check that transcription is accurate and complete
• Node 9 (Detect Language): Confirm the correct language code is identified
• Node 10 (Translate Text): Verify translation quality and accuracy
• Node 11 (Build Slack Message): Review the formatted message structure
• Node 12 (Send to Slack): Confirm the message appears in your Slack channel

Step 3: Expected Results and Success Indicators

✅ Success looks like:

• Voice message transcribed accurately in original language
• Language correctly identified
• Translation provided in target language
• Formatted message appears in Slack within 5-10 seconds
• All metadata (duration, timestamp) included
• No errors in the execution log

🔍 Common issues to check:

• Telegram file URL returns 404 → Verify Chat ID and Bot Token
• Transcription is garbled → Check audio quality and format
• Translation is inaccurate → Review OpenAI API status and model selection
• Message doesn't appear in Slack → Verify channel permissions and bot token

Congratulations! Your voice message processing workflow is now live and ready to transform how your team communicates across languages and platforms.