RedOne This workflow corresponds to n8n.io template #4638 — we link there as the canonical source.
This workflow follows the Agent → HTTP Request recipe pattern — see all workflows that pair these two integrations.
The workflow JSON
Copy or download the full n8n JSON below. Paste it into a new n8n workflow, add your credentials, activate. Full import guide →
{
"id": "X7pMjJBOk8H2tj1W",
"meta": {
"templateCredsSetupCompleted": true
},
"name": "AI Audio Assistant",
"tags": [],
"nodes": [
{
"id": "ccbc26b7-4e64-467f-844f-287bb51d201b",
"name": "Transcribe with Deepgram",
"type": "n8n-nodes-base.httpRequest",
"position": [
1040,
-420
],
"parameters": {
"url": "https://api.deepgram.com/v1/listen?model=nova-2&language=en&smart_format=true",
"method": "POST",
"options": {
"response": {
"response": {
"fullResponse": true
}
}
},
"sendBody": true,
"contentType": "binaryData",
"sendHeaders": true,
"headerParameters": {
"parameters": [
{
"name": "Authorization",
"value": "Token {{YOUR DEEP GRAM TOKEN HERE}}"
},
{
"name": "Content-Type",
"value": "audio/ogg"
}
]
},
"inputDataFieldName": "data"
},
"typeVersion": 3
},
{
"id": "42cb24ce-0ec0-4e29-961f-6aa3271e57c0",
"name": "Download Voice File",
"type": "n8n-nodes-base.httpRequest",
"position": [
680,
-420
],
"parameters": {
"url": "=https://api.telegram.org/file/bot{{ $json.telegramToken }}/{{ $json.result.file_path }}",
"options": {}
},
"typeVersion": 3
},
{
"id": "cf10e5a6-08ce-4d6f-bda2-079e4d0cc51f",
"name": "Get File Path",
"type": "n8n-nodes-base.httpRequest",
"position": [
760,
-280
],
"parameters": {
"url": "=https://api.telegram.org/bot{{ $json.telegramToken }}/getFile?file_id={{ $json.voice_file_id }}",
"options": {}
},
"typeVersion": 4.2
},
{
"id": "220308ea-8c89-4353-9b41-976feada6d30",
"name": "Prepare Voice Message Data",
"type": "n8n-nodes-base.set",
"position": [
460,
-280
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "e0deb0df-a5f8-4d13-8af6-92d0e1f0ad74",
"name": "user_id",
"type": "number",
"value": "={{ $json.message.from.id }}"
},
{
"id": "d5b76796-10f1-41ef-bff6-1602fb434ad8",
"name": "chat_id",
"type": "number",
"value": "={{ $json.message.chat.id }}"
},
{
"id": "59bda82b-6c18-40fd-b120-ed078a5d4c96",
"name": "voice_file_id",
"type": "string",
"value": "={{ $json.message.voice.file_id }}"
},
{
"id": "9412f498-5e52-4ea9-81a3-9d76379406c0",
"name": "timestamp",
"type": "string",
"value": "={{ new Date().toISOString() }}"
},
{
"id": "telegramToken",
"name": "telegramToken",
"type": "string",
"value": "Token {{YOUR TELEGRAM TOKEN HERE}}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "8c1732d1-b1d6-492e-a1c8-ed73000fc2ac",
"name": "Prepare Text Message Data",
"type": "n8n-nodes-base.set",
"position": [
260,
100
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "text-user-id",
"name": "user_id",
"type": "number",
"value": "={{ $json.message.from.id }}"
},
{
"id": "text-chat-id",
"name": "chat_id",
"type": "number",
"value": "={{ $json.message.chat.id }}"
},
{
"id": "text-message-text",
"name": "message_text",
"type": "string",
"value": "={{ $json.message.text }}"
},
{
"id": "text-message-id",
"name": "message_id",
"type": "number",
"value": "={{ $json.message.message_id }}"
},
{
"id": "text-timestamp",
"name": "timestamp",
"type": "string",
"value": "={{ new Date().toISOString() }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "419734f9-3f9f-46de-a721-8e0ffb71ca0c",
"name": "Knowledge base",
"type": "n8n-nodes-base.supabaseTool",
"position": [
980,
320
],
"parameters": {
"tableId": "knowledge_base",
"operation": "getAll"
},
"credentials": {
"supabaseApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "3ecb0359-7f32-4cb8-a8a7-692b3430df11",
"name": "Telegram Memory",
"type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
"position": [
780,
320
],
"parameters": {
"sessionKey": "telegram_{{ $json.user_id }}",
"sessionIdType": "customKey"
},
"typeVersion": 1.3
},
{
"id": "c0bba125-8492-4bde-a8e3-7fc1733b6d9d",
"name": "OpenAI Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
580,
300
],
"parameters": {
"options": {
"maxTokens": 200,
"temperature": 0.3
}
},
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.1
},
{
"id": "c0d3076c-ae1f-4fa9-85f5-1ecbcc67a91f",
"name": "AI Agent",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
760,
100
],
"parameters": {
"text": "={{ $json.body?.results?.channels?.[0]?.alternatives?.[0]?.transcript || $json.message_text || $json.message?.text || 'Hello' }}",
"options": {
"systemMessage": "You are a helpful and professional AI assistant for voice and text interactions. Your role is to:\n\n1. Understand customer inquiries and provide accurate, helpful responses\n2. Use the knowledge base tool to find relevant information when needed\n3. Maintain a friendly, professional tone\n4. Provide clear, concise responses suitable for both voice and text delivery\n5. Always aim to provide value and resolve customer concerns\n\nImportant guidelines:\n- Be empathetic and understanding\n- Provide clear, actionable responses\n- Keep responses concise for voice delivery (under 200 words)\n- If information is not in the knowledge base, say so honestly\n- For complex technical issues, recommend escalation to human support\n\nCompany context: You represent a professional service business focused on customer satisfaction and quality support."
},
"promptType": "define"
},
"typeVersion": 1.7
},
{
"id": "763b8a6a-0a33-48a9-a18f-5098fe35e5cc",
"name": "Has Transcript?",
"type": "n8n-nodes-base.if",
"position": [
1080,
-160
],
"parameters": {
"conditions": {
"string": [
{
"value1": "={{ $json.body && $json.body.results && $json.body.results.channels && $json.body.results.channels[0] && $json.body.results.channels[0].alternatives && $json.body.results.channels[0].alternatives[0] && $json.body.results.channels[0].alternatives[0].transcript }}",
"operation": "isNotEmpty"
}
]
}
},
"typeVersion": 1
},
{
"id": "8f702d09-8595-4d79-bc7d-442f0a41052e",
"name": "Is Voice Message?",
"type": "n8n-nodes-base.if",
"position": [
240,
-140
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "876dc746-b9cb-4b7f-a0e2-1542f67fbe05",
"operator": {
"type": "boolean",
"operation": "true",
"singleValue": true
},
"leftValue": "={{ Boolean($json.message.voice) }}",
"rightValue": ""
}
]
}
},
"typeVersion": 2.2
},
{
"id": "a1747da7-7e04-4766-b9bd-b3d838f956dc",
"name": "Telegram Trigger",
"type": "n8n-nodes-base.telegramTrigger",
"position": [
-20,
-140
],
"parameters": {
"updates": [
"message"
],
"additionalFields": {}
},
"credentials": {
"telegramApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.1
},
{
"id": "69cdbf6f-ee97-494a-9f3c-a99ddc917093",
"name": "Generate Audio Response",
"type": "n8n-nodes-base.httpRequest",
"position": [
1420,
-40
],
"parameters": {
"url": "https://api.deepgram.com/v1/speak",
"method": "POST",
"options": {},
"sendBody": true,
"sendQuery": true,
"sendHeaders": true,
"bodyParameters": {
"parameters": [
{
"name": "text",
"value": "={{ $json.output }}"
}
]
},
"queryParameters": {
"parameters": [
{
"name": "model",
"value": "aura-asteria-en"
}
]
},
"headerParameters": {
"parameters": [
{
"name": "Authorization",
"value": "Token {{YOUR DEEP GRAM TOKEN HERE}}"
},
{
"name": "Content-Type",
"value": "application/json"
}
]
}
},
"typeVersion": 4.2
},
{
"id": "e67b288f-1946-430b-9a45-0e62283c1d6b",
"name": "Send Audio Response",
"type": "n8n-nodes-base.telegram",
"position": [
1680,
-80
],
"parameters": {
"chatId": "={{ $('Prepare Voice Message Data').item.json.chat_id }}",
"operation": "sendAudio",
"binaryData": true,
"additionalFields": {
"caption": "\ud83c\udfa4 Voice response from AI Assistant"
}
},
"credentials": {
"telegramApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.2
},
{
"id": "8e409205-cad6-41a9-81cf-177c054a7b12",
"name": "Workflow Guide",
"type": "n8n-nodes-base.stickyNote",
"position": [
-580,
-500
],
"parameters": {
"color": 6,
"width": 360,
"height": 1160,
"content": "## \ud83c\udf99\ufe0f VoiceFlow AI: Telegram + Deepgram + OpenAI + Supabase Audio Assistant\n\n**Who is this for?**\nBusinesses, customer service teams, content creators, and organizations who want to provide intelligent voice-based interactions through Telegram. Perfect for accessibility-focused services, multilingual support, or hands-free customer assistance.\n\n**What problem does this solve?**\n- Enables natural voice conversations with AI\n- Breaks down language and accessibility barriers \n- Provides instant voice responses to customer queries\n- Reduces typing requirements for users\n- Offers 24/7 voice-based customer support\n- Maintains conversation context across voice interactions\n- Handles both voice and text messages seamlessly\n\n**What this workflow does:**\n1. Receives voice OR text messages via Telegram bot\n2. For voice: Transcribes audio using Deepgram's speech-to-text API\n3. Processes queries through AI agent with knowledge base access\n4. Generates intelligent responses based on conversation context\n5. For voice input: Converts AI response to natural speech using Deepgram TTS\n6. Sends appropriate response (audio for voice, text for text) back to user\n7. Maintains conversation memory for contextual interactions\n\n**Setup Instructions:**\n1. Get Telegram bot token from @BotFather\n2. Sign up for Deepgram API (STT + TTS services)\n3. Configure OpenAI API key for AI responses\n4. Set up Supabase knowledge base table\n5. Update API credentials in workflow nodes\n6. Test with both voice and text messages\n\n**Key Features:**\n- Dual input support (voice and text)\n- Intelligent response routing\n- Voice-to-voice conversation capability\n- Knowledge base integration\n- Conversation memory\n- Professional AI responses"
},
"typeVersion": 1
},
{
"id": "f0600a4a-5671-4cbf-9769-b7a3fcb64474",
"name": "Was Voice Input?",
"type": "n8n-nodes-base.if",
"position": [
1120,
100
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "voice-input-check",
"operator": {
"type": "boolean",
"operation": "true",
"singleValue": true
},
"leftValue": "={{ Boolean($('Telegram Trigger').item.json.message.voice) }}",
"rightValue": ""
}
]
}
},
"typeVersion": 2.2
},
{
"id": "d0d0d41a-fa55-4e6b-8b7f-c52904af284e",
"name": "Send Text Response",
"type": "n8n-nodes-base.telegram",
"position": [
1460,
180
],
"parameters": {
"text": "={{ $json.output }}",
"chatId": "={{ $('Prepare Text Message Data').item.json.chat_id }}",
"additionalFields": {
"parse_mode": "Markdown"
}
},
"credentials": {
"telegramApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.2
}
],
"active": false,
"settings": {
"executionOrder": "v1"
},
"versionId": "59537604-b414-4298-9408-8dfbcdf8e213",
"connections": {
"AI Agent": {
"main": [
[
{
"node": "Was Voice Input?",
"type": "main",
"index": 0
}
]
]
},
"Get File Path": {
"main": [
[
{
"node": "Download Voice File",
"type": "main",
"index": 0
}
]
]
},
"Knowledge base": {
"ai_tool": [
[
{
"node": "AI Agent",
"type": "ai_tool",
"index": 0
}
]
]
},
"Has Transcript?": {
"main": [
[
{
"node": "AI Agent",
"type": "main",
"index": 0
}
]
]
},
"Telegram Memory": {
"ai_memory": [
[
{
"node": "AI Agent",
"type": "ai_memory",
"index": 0
}
]
]
},
"Telegram Trigger": {
"main": [
[
{
"node": "Is Voice Message?",
"type": "main",
"index": 0
}
]
]
},
"Was Voice Input?": {
"main": [
[
{
"node": "Generate Audio Response",
"type": "main",
"index": 0
}
],
[
{
"node": "Send Text Response",
"type": "main",
"index": 0
}
]
]
},
"Is Voice Message?": {
"main": [
[
{
"node": "Prepare Voice Message Data",
"type": "main",
"index": 0
}
],
[
{
"node": "Prepare Text Message Data",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Chat Model": {
"ai_languageModel": [
[
{
"node": "AI Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Download Voice File": {
"main": [
[
{
"node": "Transcribe with Deepgram",
"type": "main",
"index": 0
}
]
]
},
"Generate Audio Response": {
"main": [
[
{
"node": "Send Audio Response",
"type": "main",
"index": 0
}
]
]
},
"Transcribe with Deepgram": {
"main": [
[
{
"node": "Has Transcript?",
"type": "main",
"index": 0
}
]
]
},
"Prepare Text Message Data": {
"main": [
[
{
"node": "AI Agent",
"type": "main",
"index": 0
}
]
]
},
"Prepare Voice Message Data": {
"main": [
[
{
"node": "Get File Path",
"type": "main",
"index": 0
}
]
]
}
}
}
Credentials you'll need
Each integration node will prompt for credentials when you import. We strip credential IDs before publishing — you'll add your own.
openAiApisupabaseApitelegramApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
Who is this for? Businesses, customer service teams, content creators, and organizations who want to provide intelligent voice-based interactions through Telegram. Perfect for accessibility-focused services, multilingual support, or hands-free customer assistance.
Source: https://n8n.io/workflows/4638/ — original creator credit. Request a take-down →
Related workflows
Workflows that share integrations, category, or trigger type with this one. All free to copy and import.
Digital marketers, content creators, social media managers, and businesses who want to use AI marketing automation for YouTube Shorts without spending hours on production. This AI workflow helps anyon
OpenAI, HTTP Request, OpenAI Chat +7
This automation is designed to help you generate AI-powered music tracks, cover art, and fully rendered music videos — all triggered from a simple Telegram chat and managed via Google Sheets.
OpenAI Chat, Memory Buffer Window, Output Parser Structured +11
Template Carnaval - time instagram. Uses toolWorkflow, lmChatOpenAi, memoryBufferWindow, agent. Event-driven trigger; 56 nodes.
Tool Workflow, OpenAI Chat, Memory Buffer Window +10
This workflow contains community nodes that are only compatible with the self-hosted version of n8n.
Output Parser Structured, Telegram, N8N Nodes Tesseractjs +14
Multi-modal AI Image Generator powered by Google's Nano Banana (Gemini 2.5 Flash Image)* - the latest state-of-the-art image generation model Accepts text, images, voice messages, and PDFs via Telegra
OpenAI, HTTP Request, Telegram +5