Voice Message Translator Bridge: Telegram to Slack - n8n Workflow

Teams using Telegram for quick communication who need to bridge language gaps in Slack.
Users managing international communities requiring automatic voice translation.
n8n developers looking for advanced examples of binary file processing and chained AI calls.
Anyone interested in leveraging the OpenAI Whisper model within an n8n workflow.

Managing communications across multiple language groups can be challenging, especially when using voice messages. This sophisticated n8n workflow solves this by creating a real-time, automated translation bridge. When a user sends a voice message in Telegram (such as Japanese or English), this n8n automation detects the source language, uses the OpenAI Whisper model to transcribe the audio, and then uses GPT-4o-mini for rapid translation. The final output—including the original transcript, the translation, and the sender's details—is formatted neatly and posted to a Slack channel. This integration demonstrates the power of the n8n node ecosystem for creating complex, multi-step AI pipelines, making it a valuable addition to your collection of n8n templates.

This n8n workflow begins with the specialized Telegram n8n trigger.

1. Trigger & Filtering: The workflow is activated by the Telegram n8n trigger whenever a new message is received. An If n8n node checks if the message contains a voice element, filtering out all other message types.


2. File Retrieval Setup: The workflow captures the Telegram fileid and bot token, then makes an HTTP request to the Telegram API to retrieve the actual filepath. A Function n8n node then constructs the complete, downloadable file URL.


3. Download & Transcription: An HTTP Request n8n node downloads the voice data as a binary file. A subsequent Code n8n node prepares the binary data for the AI service. The Transcribe a recording OpenAI n8n node (using Whisper) converts the binary audio file into text.


4. Language Detection & Translation: The transcribed text is passed to a Function n8n node (Detect Language) which determines if the input is Japanese or English based on character sets. This sets the source and target languages. The Translate (OpenAI) HTTP Request n8n node calls the GPT-4o-mini model, providing the transcribed text and language instructions to get the translated result.


5. Slack Posting: The Build Slack Message Function n8n node formats the translation, adding user context (Telegram username) and language flags. Finally, the Post to Slack HTTP Request n8n node publishes the complete, translated message to the designated Slack channel using a Slack bot token. This entire process is a seamless n8n workflow execution.

To deploy this powerful n8n workflow template, follow these steps:

1. Import the n8n Workflow: Copy the provided JSON data and import it directly into your n8n instance using the 'New' > 'Import from JSON' option.


2. Telegram Credentials: Set up the Telegram Trigger n8n node. You will need a Telegram Bot token. Create a new credential (TELEGRAMBOTCRED) or update the existing placeholder.


3. OpenAI Credentials: This n8n workflow requires two separate HTTP Header Auth credentials for OpenAI (one for Whisper transcription and one for GPT translation). They can use the same API key:

1. Slack Credentials: Create an HTTP Header Auth credential (SLACKBOTTOKENHEADER) containing your Slack Bot Token (xoxb-...). This token must have permissions (chat:write, files:read, files:write, groups:read, channels:read).


2. Configuration Nodes:

1. Activate: Save and activate the n8n workflow. It is now ready to respond to Telegram voice messages.

This n8n workflow relies on a complex chain of highly configured n8n nodes:

Telegram Trigger: The starting n8n trigger. It listens specifically for 'message' updates from your Telegram bot.
Is Voice? (If): A core logic n8n node that acts as a filter, ensuring the n8n workflow only proceeds if the incoming message contains a voice element, optimizing resource usage.
Telegram getFile (HTTP Request): Performs the necessary API call to Telegram using the file ID to retrieve the filepath required for the actual audio download.
Download Voice File (HTTP Request): This n8n node is configured with Response Format: file to handle the incoming audio data as binary, making it available for Whisper.
Prepare Whisper Input1 (Code): A critical custom code n8n node that renames the downloaded binary key from data to audio to meet the input requirements of the OpenAI transcription node.
Transcribe a recording (OpenAI n8n node): This specialized n8n node uses the OpenAI Whisper model to convert the binary audio file into readable text.
Detect Language (Function): Uses JavaScript regex to quickly determine the source language (Japanese or English) of the transcription, setting the stage for accurate translation by the next n8n node.
Translate (OpenAI) (HTTP Request): Directly interacts with the GPT-4o-mini API via a custom n8n node configuration, instructing the model to perform translation based on the detected language pair (sourcelang and targetlang).
Build Slack Message (Function): A final Function n8n node that compiles all gathered data—username, original transcript, translation, and language flags—into the final, readable slacktext string.
Post to Slack (HTTP Request): The concluding n8n node, responsible for sending the formatted slack_text to the target Slack channel.