Automatisez vos appels vocaux avec la synthèse vocale
Ce workflow n8n vous permet de transformer des messages texte en appels vocaux automatisés grâce à l'API ClickSend. I...
Documentation Complète
📋 Chatbot Telegram AI Multi-Format Avancé
💡 Description
Optimisez l'interaction client avec notre chatbot Telegram intelligent, capable de comprendre et de répondre aux messages texte et vocaux. Grâce à l'intégration d'un modèle OpenAI, ce workflow automatise les réponses personnalisées tout en corrigeant les erreurs potentielles dans la communication, offrant ainsi une expérience utilisateur fluide et engageante. Idéal pour les entreprises cherchant à améliorer leur service client et à automatiser les processus de communication, réduisant ainsi le temps passé sur des tâches répétitives.
📈 Impact & ROI: En améliorant l'efficacité du service client avec des réponses automatisées précises, les entreprises peuvent réduire les coûts opérationnels tout en augmentant la satisfaction client, conduisant potentiellement à une fidélisation accrue et un meilleur retour sur investissement.
🚀 Fonctionnalités Clés
- ✅ Réponses intelligentes et personnalisées avec OpenAI
- ✅ Support multi-format pour texte et voix
- ✅ Amélioration continue grâce à la correction automatique des erreurs
- ✅ Automatisation avancée pour un engagement client optimisé
📊 Architecture Technique
15
Nodes
Nodes
9
Connexions
Connexions
2
Services
Services
🔌 Services Intégrés
TelegramOpenAI
🔧 Composition du Workflow
| Node | Type | Description |
|---|---|---|
| OpenAI Chat Model | @n8n/n8n-nodes-langchain.lmChatOpenAi | Traitement des données |
| Window Buffer Memory | @n8n/n8n-nodes-langchain.memoryBufferWindow | Traitement des données |
| Correct errors | telegram | Traitement des données |
| Listen for incoming events | telegramTrigger | Traitement des données |
| Download voice file | telegram | Traitement des données |
| Combine content and set properties | set | Traitement des données |
| Send final reply | telegram | Traitement des données |
| Send error message | telegram | Traitement des données |
| Convert audio to text | @n8n/n8n-nodes-langchain.openAi | Traitement des données |
| Sticky Note | stickyNote | Traitement des données |
| Sticky Note1 | stickyNote | Traitement des données |
| Sticky Note2 | stickyNote | Traitement des données |
| Send Typing action | telegram | Traitement des données |
| AI Agent | @n8n/n8n-nodes-langchain.agent | Traitement des données |
| Determine content type | switch | Traitement des données |
📖 Guide d'Implémentation
- Import du workflow: Téléchargez le fichier JSON et importez-le dans votre instance n8n
- Configuration des credentials: Configurez les accès pour chaque service utilisé
- Personnalisation: Adaptez les paramètres selon vos besoins spécifiques
- Test: Exécutez le workflow en mode test pour vérifier le bon fonctionnement
- Activation: Activez le workflow pour une exécution automatique
🏷️ Tags
Structure JSON
Voir le code JSON complet
{
"id": "HJwTWtzlhK8Q5SOv",
"meta": {
"instanceId": "fb924c73af8f703905bc09c9ee8076f48c17b596ed05b18c0ff86915ef8a7c4a",
"templateCredsSetupCompleted": true
},
"name": "Telegram AI multi-format chatbot",
"tags": [],
"nodes": [
{
"id": "65196267-0d57-4af4-9081-962701478146",
"name": "OpenAI Chat Model",
"type": "@n8n\/n8n-nodes-langchain.lmChatOpenAi",
"position": [
660,
640
],
"parameters": {
"model": "gpt-4o",
"options": {
"temperature": 0.7,
"frequencyPenalty": 0.2
}
},
"credentials": {
"openAiApi": {
"id": "rveqdSfp7pCRON1T",
"name": "Ted's Tech Talks OpenAi"
}
},
"typeVersion": 1
},
{
"id": "fc446ef0-2f15-42e7-a993-7960d76d8876",
"name": "Window Buffer Memory",
"type": "@n8n\/n8n-nodes-langchain.memoryBufferWindow",
"position": [
800,
640
],
"parameters": {
"sessionKey": "=chat_with_{{ $('Listen for incoming events').first().json.message.chat.id }}",
"contextWindowLength": 10
},
"typeVersion": 1
},
{
"id": "51c3cddd-fc21-4fff-b615-ea7080c47947",
"name": "Correct errors",
"type": "n8n-nodes-base.telegram",
"position": [
1220,
580
],
"parameters": {
"text": "={{ $('AI Agent').item.json.output.replace(\/&\/g, \"&\").replace(\/>\/g, \">\").replace(\/<\/g, \"<\").replace(\/\"\/g, \""\") }}",
"chatId": "={{ $('Listen for incoming events').first().json.message.from.id }}",
"additionalFields": {
"parse_mode": "HTML",
"appendAttribution": false
}
},
"credentials": {
"telegramApi": {
"id": "9dexJXnlVPA6wt8K",
"name": "Chat & Sound"
}
},
"typeVersion": 1.1
},
{
"id": "d931b7e1-bc17-431e-ae67-967b6ef79236",
"name": "Listen for incoming events",
"type": "n8n-nodes-base.telegramTrigger",
"position": [
-440,
480
],
"webhookId": "322dce18-f93e-4f86-b9b1-3305519b7834",
"parameters": {
"updates": [
"*"
],
"additionalFields": []
},
"credentials": {
"telegramApi": {
"id": "9dexJXnlVPA6wt8K",
"name": "Chat & Sound"
}
},
"typeVersion": 1
},
{
"id": "b33335ff-5dea-4fff-8f63-fea2b11b8241",
"name": "Download voice file",
"type": "n8n-nodes-base.telegram",
"position": [
60,
600
],
"parameters": {
"fileId": "={{$json.message.voice.file_id}}",
"resource": "file"
},
"credentials": {
"telegramApi": {
"id": "9dexJXnlVPA6wt8K",
"name": "Chat & Sound"
}
},
"typeVersion": 1.2
},
{
"id": "2954ced6-ab98-42e6-bf64-237146a433e0",
"name": "Combine content and set properties",
"type": "n8n-nodes-base.set",
"position": [
440,
460
],
"parameters": {
"options": [],
"assignments": {
"assignments": [
{
"id": "bccbce0a-7786-49c9-979a-7a285cb69f78",
"name": "CombinedMessage",
"type": "string",
"value": "={{ $json.message && $json.message.text ? $json.message.text : ($json.text ? $json.text : '') }}"
},
{
"id": "5b1fc9f5-1408-4099-88cc-a23725c9eddb",
"name": "Message Type ",
"type": "string",
"value": "={{ $json?.message?.text && !$json?.text ? \"text query\" : (!$json?.message?.text && $json?.text ? \"voice message\" : \"unknown type message\") }}"
},
{
"id": "1e9a17fa-ec5d-49dc-9ff6-1f28b57fb02e",
"name": "Source Type",
"type": "string",
"value": "={{ $('Listen for incoming events').item.json.message.forward_origin ? \" forwarded\" : \"\" }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "e18de374-941f-4c2e-ab6c-6c6f68f2ce12",
"name": "Send final reply",
"type": "n8n-nodes-base.telegram",
"onError": "continueErrorOutput",
"position": [
1040,
460
],
"parameters": {
"text": "={{ $json.output }} \n\nThank you for your{{ $('Combine content and set properties').item.json['Source Type'] }} {{ $('Combine content and set properties').item.json['Message Type '] }} 🤗",
"chatId": "={{ $('Listen for incoming events').first().json.message.from.id }}",
"additionalFields": {
"parse_mode": "HTML",
"appendAttribution": false
}
},
"credentials": {
"telegramApi": {
"id": "9dexJXnlVPA6wt8K",
"name": "Chat & Sound"
}
},
"typeVersion": 1.1
},
{
"id": "b47a9583-ce5c-464f-a9e6-153fb42e685f",
"name": "Send error message",
"type": "n8n-nodes-base.telegram",
"position": [
60,
300
],
"parameters": {
"text": "=Sorry, {{ $('Listen for incoming events').first().json.message.from.first_name }}! This command is not supported yet. Please send text or voice messages.",
"chatId": "={{ $('Listen for incoming events').first().json.message.from.id }}",
"additionalFields": {
"parse_mode": "Markdown",
"appendAttribution": false
}
},
"credentials": {
"telegramApi": {
"id": "9dexJXnlVPA6wt8K",
"name": "Chat & Sound"
}
},
"typeVersion": 1.2
},
{
"id": "0196b49e-90a1-4f2f-8b94-492fced37dbf",
"name": "Convert audio to text",
"type": "@n8n\/n8n-nodes-langchain.openAi",
"position": [
240,
600
],
"parameters": {
"options": {
"language": "",
"temperature": 0.7
},
"resource": "audio",
"operation": "transcribe"
},
"credentials": {
"openAiApi": {
"id": "rveqdSfp7pCRON1T",
"name": "Ted's Tech Talks OpenAi"
}
},
"typeVersion": 1.5
},
{
"id": "66505b83-e0c3-4d9d-8e1a-9b54030e29e7",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
-466.12784869794086,
220
],
"parameters": {
"width": 1035.4478381373049,
"height": 547.5630890194532,
"content": "## Receive and pre-process messages \n"
},
"typeVersion": 1
},
{
"id": "44087d3f-86c8-407c-8791-645d167165cb",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
620,
220
],
"parameters": {
"color": 2,
"width": 861.262180151035,
"height": 550.5748478134515,
"content": "## 1. Send incoming message to the AI Agent\n## 2. Deliver agent reply to the user \n"
},
"typeVersion": 1
},
{
"id": "d7e58831-de97-483f-8b8a-583f85397245",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
20,
553.0639243489702
],
"parameters": {
"color": 6,
"width": 367.73614918993235,
"height": 194.83713159725437,
"content": "## Transcribe audio"
},
"typeVersion": 1
},
{
"id": "89515d80-6efc-40a8-95ce-343d4ff4dbee",
"name": "Send Typing action",
"type": "n8n-nodes-base.telegram",
"position": [
-180,
300
],
"parameters": {
"chatId": "={{ $('Listen for incoming events').first().json.message.from.id }}",
"operation": "sendChatAction"
},
"credentials": {
"telegramApi": {
"id": "9dexJXnlVPA6wt8K",
"name": "Chat & Sound"
}
},
"typeVersion": 1.2
},
{
"id": "c925d059-f843-473c-bfd4-3c563d80ca0f",
"name": "AI Agent",
"type": "@n8n\/n8n-nodes-langchain.agent",
"position": [
680,
460
],
"parameters": {
"text": "={{ $json.CombinedMessage }}",
"options": {
"humanMessage": "TOOLS\n------\nAssistant can ask the user to use tools to look up information that may be helpful in answering the users original question. The tools the human can use are:\n\n{tools}\n\n{format_instructions}\n\nUSER'S INPUT\n--------------------\nHere is the user's input (remember to respond with a markdown code snippet of a json blob with a single action, and NOTHING else):\n\n{{input}}",
"systemMessage": "=You are a helpful AI assistant. You are chatting with the user named `{{ $('Determine content type').item.json.message.from.first_name }}`. You need to address the user by their name. Today is {{ DateTime.fromISO($now).toLocaleString(DateTime.DATETIME_FULL) }}\n\nIn your reply, always send a message in Telegram-supported HTML format. Here are the formatting instructions:\n1. The following tags are currently supported:\n<b>bold<\/b>, <strong>bold<\/strong>\n<i>italic<\/i>, <em>italic<\/em>\n<u>underline<\/u>, <ins>underline<\/ins>\n<s>strikethrough<\/s>, <strike>strikethrough<\/strike>, <del>strikethrough<\/del>\n<span class=\"tg-spoiler\">spoiler<\/span>, <tg-spoiler>spoiler<\/tg-spoiler>\n<b>bold <i>italic bold <s>italic bold strikethrough <span class=\"tg-spoiler\">italic bold strikethrough spoiler<\/span><\/s> <u>underline italic bold<\/u><\/i> bold<\/b>\n<a href=\"http:\/\/www.example.com\/\">inline URL<\/a>\n<code>inline fixed-width code<\/code>\n<pre>pre-formatted fixed-width code block<\/pre>\n2. Any code that you send should be wrapped in these tags: <pre><code class=\"language-python\">pre-formatted fixed-width code block written in the Python programming language<\/code><\/pre>\nOther programming languages are supported as well.\n3. All <, > and & symbols that are not a part of a tag or an HTML entity must be replaced with the corresponding HTML entities (< with <, > with > and & with &)\n4. If the user sends you a message starting with \/ sign, it means this is a Telegram bot command. For example, all users send \/start command as their first message. Try to figure out what these commands mean and reply accodringly\n"
}
},
"typeVersion": 1.1
},
{
"id": "2c56536d-1a86-4a49-b495-3e877adb308a",
"name": "Determine content type",
"type": "n8n-nodes-base.switch",
"position": [
-180,
480
],
"parameters": {
"rules": {
"values": [
{
"outputKey": "Text",
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"operator": {
"type": "string",
"operation": "notEmpty",
"singleValue": true
},
"leftValue": "={{ $json.message.text }}",
"rightValue": "\/"
}
]
},
"renameOutput": true
},
{
"outputKey": "Voice",
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "dd41bbf0-bee0-450b-9160-b769821a4abc",
"operator": {
"type": "object",
"operation": "exists",
"singleValue": true
},
"leftValue": "={{ $json.message.voice}}",
"rightValue": ""
}
]
},
"renameOutput": true
}
]
},
"options": {
"fallbackOutput": "extra"
}
},
"typeVersion": 3.2
}
],
"active": false,
"pinData": [],
"settings": {
"executionOrder": "v1"
},
"versionId": "15ae799b-6868-4519-b579-3f202e4de5b2",
"connections": {
"AI Agent": {
"main": [
[
{
"node": "Send final reply",
"type": "main",
"index": 0
}
]
]
},
"Send final reply": {
"main": [
[],
[
{
"node": "Correct errors",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Chat Model": {
"ai_languageModel": [
[
{
"node": "AI Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Download voice file": {
"main": [
[
{
"node": "Convert audio to text",
"type": "main",
"index": 0
}
]
]
},
"Window Buffer Memory": {
"ai_memory": [
[
{
"node": "AI Agent",
"type": "ai_memory",
"index": 0
}
]
]
},
"Convert audio to text": {
"main": [
[
{
"node": "Combine content and set properties",
"type": "main",
"index": 0
}
]
]
},
"Determine content type": {
"main": [
[
{
"node": "Combine content and set properties",
"type": "main",
"index": 0
}
],
[
{
"node": "Download voice file",
"type": "main",
"index": 0
}
],
[
{
"node": "Send error message",
"type": "main",
"index": 0
}
]
]
},
"Listen for incoming events": {
"main": [
[
{
"node": "Determine content type",
"type": "main",
"index": 0
},
{
"node": "Send Typing action",
"type": "main",
"index": 0
}
]
]
},
"Combine content and set properties": {
"main": [
[
{
"node": "AI Agent",
"type": "main",
"index": 0
}
]
]
}
}
}
Workflows Similaires
Automatisez vos Emails avec l'Agent IA Professionnel
Optimisez la gestion de vos emails grâce à ce workflow d'automatisation avancé. Conçu pour les entreprises cherchant...
Surveillez les Mentions Twitter avec RocketChat
Ce workflow automatise la surveillance des mentions Twitter de votre marque, en l'occurrence @n8n_io, et vous informe in...