Automated Professional Email Drafting from Telegram Voice Notes - n8n Workflow

Professionals who frequently use voice notes for communication.
Users looking for high-quality, reliable AI generation within an n8n workflow.
n8n developers interested in advanced LangChain agent implementations.
Anyone seeking efficient, hands-free email drafting solutions that utilize the power of n8n templates.

This advanced n8n workflow solves the common problem of converting informal voice messages into formal, ready-to-send professional correspondence. The process begins with the critical Telegram Voice Message n8n trigger, which listens for new voice notes. It seamlessly integrates OpenAI's powerful models: first using Whisper for highly accurate transcription, and then employing a specialized GPT-4 AI Agent. This agent is configured within the n8n node settings to not only understand the intent but also to generate the final email content, complete with HTML formatting and recipient detection. By leveraging the Structured Output Parser n8n node, the data integrity is guaranteed, ensuring the resulting email is perfectly formatted and ready to be sent using the Gmail n8n node. This end-to-end n8n workflow drastically improves communication efficiency.

This automation follows a precise multi-stage flow to ensure accuracy and professionalism:

1. Trigger and Download: The n8n workflow is initiated by the Telegram Voice Message Trigger n8n trigger when a voice note is received. The flow then proceeds to use a Telegram n8n node to download the associated audio file.


2. Transcription (Whisper): The Transcribe Audio to Text (OpenAI Whisper) n8n node processes the downloaded audio, converting the spoken content into raw text.


3. AI Agent Logic: The core processing happens in the Generate Email Content AI AGENT n8n node. This n8n node receives the transcription and executes a highly detailed, custom prompt instructing the agent to detect the email recipient, determine the purpose of the message (e.g., resignation, request), and generate a professional subject line and a fully formatted HTML body.


4. Model and Parsing: The AI Agent utilizes the GPT-4 Email Generator Model (specifically, gpt-4.1-mini) and is strictly governed by the Enforce Email JSON Schema n8n node. This ensures the output is a guaranteed JSON object containing the email, subject, and body keys.


5. Final Delivery (Gmail): Finally, the Send Email Gmail n8n node maps the structured data from the AI output (using expressions like {{ $json.output.email }}) to draft and send the polished, professional email to the intended recipient.

To implement this powerful n8n workflow, follow these steps:

1. Import: Copy the provided n8n workflow JSON data and paste it into your n8n instance using the 'New' -> 'Import from JSON' function.


2. Telegram Setup: Set up credentials for the Telegram Voice Message Trigger n8n trigger, linking it to your desired Telegram Bot (via BotFather).


3. OpenAI Credentials: Configure your OpenAI API Key for the Transcribe Audio to Text (OpenAI Whisper) and the associated AI Agent nodes. Ensure your API key has access to the Whisper transcription service and the specified GPT-4 model.


4. Gmail Credentials: Set up your OAuth2 credentials for the Gmail n8n node used for sending the final email.


5. Activation: Once all credentials are configured, activate the n8n workflow to start listening for incoming Telegram voice messages. This n8n templates is now ready for use.

Telegram Voice Message Trigger (n8n trigger): Initiates the n8n workflow upon receiving a message update in the Telegram bot. Configured specifically to look for messages.
Download Telegram Voice File (Telegram n8n node): Responsible for retrieving the actual audio file from Telegram's servers using the dynamically provided file_id found in the initial trigger data.
Transcribe Audio to Text (OpenAI Whisper n8n node): Uses the OpenAI API, configured for the audio resource and transcribe operation. This n8n node is essential for converting the audio into usable text for the AI.
Generate Email Content AI AGENT (Langchain n8n node): The brain of the operation. This agent is given a detailed French-language prompt to perform multiple complex reasoning tasks (intent classification, recipient parsing, content generation) and is configured to enforce a strict JSON output structure.
Enforce Email JSON Schema (Structured Output Parser n8n node): Crucial for reliability. This n8n node uses a predefined JSON schema (email, subject, body) to validate and structure the AI's output, guaranteeing reliable data transmission to the final sending step.
Send Email (Gmail n8n node): The final action. It dynamically maps the parsed email, subject, and HTML body from the AI Agent's output to send the professional message using the configured Gmail account.