创建使用OpenAI GPT4o-mini转录的语音转文本API
中级
这是一个Document Extraction, Multimodal AI领域的自动化工作流，包含 10 个节点。主要使用 Set, Webhook, HttpRequest, RespondToWebhook 等节点。 创建使用OpenAI GPT4o-mini转录的语音转文本API
前置要求
- •HTTP Webhook 端点（n8n 会自动生成）
- •可能需要目标 API 的认证凭证
工作流预览
可视化展示节点连接关系，支持缩放和平移
导出工作流
复制以下 JSON 配置到 n8n 导入，即可使用此工作流
{
"nodes": [
{
"id": "6b7dd876-ed21-47f6-877b-d6c45f8bc9b3",
"name": "使用 OpenAI 转录",
"type": "n8n-nodes-base.httpRequest",
"position": [
560,
140
],
"parameters": {
"url": "https://api.openai.com/v1/audio/transcriptions",
"method": "POST",
"options": {},
"sendBody": true,
"contentType": "multipart-form-data",
"authentication": "predefinedCredentialType",
"bodyParameters": {
"parameters": [
{
"name": "file",
"parameterType": "formBinaryData",
"inputDataFieldName": "audio_file"
},
{
"name": "model",
"value": "gpt-4o-mini-transcribe"
}
]
},
"nodeCredentialType": "openAiApi"
},
"credentials": {
"openAiApi": {
"id": "dMiSy27YCK6c6rra",
"name": "Duv's OpenAI"
}
},
"typeVersion": 4.2
},
{
"id": "26543502-9e91-4d70-af12-df78ac5ba630",
"name": "提取转录文本",
"type": "n8n-nodes-base.set",
"position": [
840,
140
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "51b3d82e-6ef0-4b0b-86aa-33cf8203a24e",
"name": "Transcript",
"type": "string",
"value": "={{ $json.text }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "d12de568-f2b8-4757-b45f-f79bc579ee36",
"name": "包含要转录的音频的 Webhook",
"type": "n8n-nodes-base.webhook",
"position": [
220,
140
],
"webhookId": "6a06f5e4-9105-4780-9840-9b7619a25647",
"parameters": {
"path": "audio-to-transcribe",
"options": {},
"httpMethod": "POST",
"responseMode": "responseNode"
},
"typeVersion": 2
},
{
"id": "e5de7e17-64a7-4466-a381-0dfb2e9d9711",
"name": "使用转录文本响应 Webhook",
"type": "n8n-nodes-base.respondToWebhook",
"position": [
1100,
140
],
"parameters": {
"options": {
"responseCode": 200
}
},
"typeVersion": 1.3
},
{
"id": "883dbfbe-7330-41e6-bc7e-6dda8385250c",
"name": "便签",
"type": "n8n-nodes-base.stickyNote",
"position": [
120,
700
],
"parameters": {
"color": 4,
"width": 580,
"height": 120,
"content": "<!DOCTYPE html>"
},
"typeVersion": 1
},
{
"id": "9c06f4c8-ae6e-43a9-9eda-a1452d81e17f",
"name": "便签1",
"type": "n8n-nodes-base.stickyNote",
"position": [
-320,
-40
],
"parameters": {
"width": 380,
"height": 860,
"content": "## Speech Transcription API Endpoint\n\nThis workflow exposes a webhook that transcribes any audio file sent to it.\n\n**How to use**\n\n1. **Add credentials:** Select the **Transcribe with OpenAI** node and add your OpenAI API key.\n2. **Get your endpoint URL:**\n\n * Make sure the workflow is **Active**.\n\n * Click the **Webhook** node and copy the **Production URL**.\n\n3. **Connect the frontend:**\n\n * Find the sticky note labeled \"Example Frontend Code Below\". Copy the code from the note beneath it.\n\n * In the code, replace the `YOUR WEBHOOK URL` placeholder with the URL you copied in step 2.\n\n\nThe provided snippet below is a great starting point. Feel free to adapt it and build the interface you need!\n\n"
},
"typeVersion": 1
},
{
"id": "b92cc8ab-64c9-4b24-a222-aa542b4bb710",
"name": "便签2",
"type": "n8n-nodes-base.stickyNote",
"position": [
120,
380
],
"parameters": {
"color": 4,
"width": 580,
"height": 300,
"content": "## Example Frontend Code Below**\nThe sticky note directly below contains a complete HTML file that creates a functional audio recording interface. It's a plug-and-play example to demonstrate how to call this webhook.\n\n**To use it:**\n\n1. Copy the entire code block from the note below.\n2. Save it as an `.html` file.\n3. **Remember to replace** the `YOUR WEBHOOK URL` placeholder inside the code with your actual URL from the Webhook node.\n4. Open the file in your browser to test."
},
"typeVersion": 1
},
{
"id": "883ba3ee-2a32-477f-8493-da931847a9cb",
"name": "便签3",
"type": "n8n-nodes-base.stickyNote",
"position": [
120,
-40
],
"parameters": {
"color": 7,
"width": 280,
"height": 360,
"content": "## The webhook to call from your app\nPOST the audio as \"audio_file\" to this webhook to start the workflow."
},
"typeVersion": 1
},
{
"id": "40e86f5a-c472-4801-9235-3a2f8e3b0088",
"name": "便签4",
"type": "n8n-nodes-base.stickyNote",
"position": [
460,
-40
],
"parameters": {
"color": 7,
"width": 280,
"height": 360,
"content": "## AI transcription with OpenAI GPT4o-mini transcribe"
},
"typeVersion": 1
},
{
"id": "4d9d11f7-ebfa-4277-bf41-9070b6d052b1",
"name": "便利贴5",
"type": "n8n-nodes-base.stickyNote",
"position": [
1020,
-20
],
"parameters": {
"color": 7,
"width": 280,
"height": 340,
"content": "## Sending the transcript back to your app\nYour app should expect the key \"Transcript\" in the body of the webhook response."
},
"typeVersion": 1
}
],
"connections": {
"Extract transcript": {
"main": [
[
{
"node": "Respond to Webhook with transcript",
"type": "main",
"index": 0
}
]
]
},
"Transcribe with OpenAI": {
"main": [
[
{
"node": "Extract transcript",
"type": "main",
"index": 0
}
]
]
},
"Webhook containing audio to transcribe": {
"main": [
[
{
"node": "Transcribe with OpenAI",
"type": "main",
"index": 0
}
]
]
}
}
}常见问题
如何使用这个工作流？
复制上方的 JSON 配置代码，在您的 n8n 实例中创建新工作流并选择「从 JSON 导入」，粘贴配置后根据需要修改凭证设置即可。
这个工作流适合什么场景？
中级 - 文档提取, 多模态 AI
需要付费吗？
本工作流完全免费，您可以直接导入使用。但请注意，工作流中使用的第三方服务（如 OpenAI API）可能需要您自行付费。
相关工作流推荐
多级文档审批与审计工作流
使用 Supabase 和 Gmail 自动化多级工作流的文档审批
If
Set
Form
+8
38 节点Alok Kumar
文档提取
通过Telegram使用Mistral OCR从图像和PDF中提取文本到Markdown
通过Telegram使用Mistral OCR从图像和PDF中提取文本到Markdown
If
Set
Code
+8
40 节点Rostislav
文档提取
智能PDF摘要生成器（带音频）
使用Groq AI总结PDF文档，并通过Qwen TTS转换为音频
Set
Code
Webhook
+7
17 节点Laiba
文档提取
使用 GPT-4 和 Airtable 自动记录和备份工作流
使用 GPT-4 和 Airtable 自动记录和备份工作流
If
N8n
Set
+14
38 节点Guillaume Duvernay
AI 摘要总结
使用Lookio和OpenAI GPT从知识源创建基于事实的文章
使用Lookio和OpenAI GPT从知识源创建基于事实的文章
Set
Split Out
Aggregate
+7
19 节点Guillaume Duvernay
内容创作
使用Super RAG和GPT-5从知识源创建基于事实的文章
使用Super RAG和GPT-5从知识源创建基于事实的文章
Set
Split Out
Aggregate
+7
19 节点Guillaume Duvernay
AI RAG 检索增强
