gpt-audio-mini

openaiopenai/gpt-audio-mini

API 使用指南

音频 Chat 调用

在消息内容中发送 base64 编码的 WAV/MP3 音频。模型用文本和/或音频回复——API 模式与 gpt-audio 相同，仅价格更低。

cURL

curl https://api.therouter.ai/v1/chat/completions \
  -H "Authorization: Bearer $THEROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-audio-mini",
    "modalities": ["text", "audio"],
    "audio": { "voice": "alloy", "format": "wav" },
    "messages": [
      {
        "role": "user",
        "content": [
          { "type": "text", "text": "What does this audio say?" },
          { "type": "input_audio", "input_audio": { "data": "<BASE64_AUDIO>", "format": "wav" } }
        ]
      }
    ]
  }'

流式音频对话

实时流式接收文本（转写）token，同时在最终消息中获取完整音频。

cURL

curl https://api.therouter.ai/v1/chat/completions \
  -H "Authorization: Bearer $THEROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-audio-mini",
    "modalities": ["text", "audio"],
    "audio": { "voice": "nova", "format": "wav" },
    "stream": true,
    "messages": [
      { "role": "user", "content": "Tell me a story in 30 seconds." }
    ]
  }'

工具/函数调用

gpt-audio-mini 支持带音频上下文的工具调用——使用方式与纯文本模型相同。

cURL

curl https://api.therouter.ai/v1/chat/completions \
  -H "Authorization: Bearer $THEROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-audio-mini",
    "messages": [
      {
        "role": "user",
        "content": [
          { "type": "input_audio", "input_audio": { "data": "<BASE64_AUDIO>", "format": "wav" } },
          { "type": "text", "text": "Extract the booking reference and the date mentioned in this audio." }
        ]
      }
    ],
    "tools": [
      {
        "type": "function",
        "function": {
          "name": "extract_booking",
          "description": "Extract booking reference and date",
          "parameters": {
            "type": "object",
            "properties": {
              "reference": { "type": "string" },
              "date": { "type": "string" }
            },
            "required": ["reference", "date"]
          }
        }
      }
    ],
    "tool_choice": "auto"
  }'

事实档案 — 本页每条断言可在此回溯来源

来源	URL	采集于
发布日期	Design for Online / OpenRouter ↗	2026-05-29	已核实
上下文窗口	OpenRouter ↗	2026-05-29	已核实
最大输出长度	OpenRouter ↗	2026-05-29	已核实
支持的模态	developers.openai.com ↗	2026-05-29	已核实
支持的能力	platform.openai.com ↗	2026-05-29	已核实
许可	—	—	已核实
Audio quality (upgraded decoder)	Design for Online ↗	2026-05-29	待核实
OpenAI 发布 gpt-audio 和 gpt-audio-mini	Design for Online / OpenRouter ↗	2026-05-29	已核实