SenseAudio 文本转语音（TTS）

SenseAudio TTS 利用大语言模型，将文本转换为自然、情感丰富的语音。支持 10+ 种情感、流式输出（SSE）以及细粒度的语音控制。

Endpoint： POST https://api.senseaudio.cn/v1/t2a_v2 Auth： Authorization: Bearer $SENSEAUDIO_API_KEY Max text length： 10,000 字符

请求参数

Headers

Body

voice_setting

audio_setting

停顿标签

在文本中插入停顿：

你好欢迎使用我们的服务

• time 单位为毫秒，最小值为 100ms

非流式响应

{
  "data": {
    "audio": "hex-encoded audio data...",
    "status": 2
  },
  "extra_info": {
    "audio_length": 3500,
    "audio_sample_rate": 32000,
    "audio_size": 56000,
    "bitrate": 128000,
    "audio_format": "mp3",
    "audio_channel": 1,
    "word_count": 24,
    "usage_characters": 30
  },
  "base_resp": {"status_code": 0, "status_msg": "success"}
}

data.audio 为十六进制编码。解码方式：bytes.fromhex(audio_hex)

SSE 流式响应

每个数据块格式为：data: {"data":{"audio":"hex...","status":1},...}

最终数据块中 status 为 2，并包含完整的 extra_info。

代码示例

curl（非流式）

curl -X POST https://api.senseaudio.cn/v1/t2a_v2 
  -H "Authorization: Bearer $SENSEAUDIO_API_KEY" 
  -H "Content-Type: application/json" 
  -d '{
    "model": "SenseAudio-TTS-1.0",
    "text": "道可道，非常道。名可名，非常名。",
    "stream": false,
    "voice_setting": {"voice_id": "male_0004_a"}
  }' -o response.json

jq -r '.data.audio' response.json | xxd -r -p > output.mp3

Python（非流式）

import requests

resp = requests.post(
    "https://api.senseaudio.cn/v1/t2a_v2",
    headers={"Authorization": "Bearer YOUR_API_KEY", "Content-Type": "application/json"},
    json={
        "model": "SenseAudio-TTS-1.0",
        "text": "道可道，非常道。",
        "stream": False,
        "voice_setting": {"voice_id": "male_0004_a"}
    }
)
result = resp.json()
audio_bytes = bytes.fromhex(result["data"]["audio"])
with open("output.mp3", "wb") as f:
    f.write(audio_bytes)

Python（SSE 流式）

import requests, json

with requests.post(
    "https://api.senseaudio.cn/v1/t2a_v2",
    headers={"Authorization": "Bearer YOUR_API_KEY", "Content-Type": "application/json"},
    json={"model": "SenseAudio-TTS-1.0", "text": "这是流式输出示例。", "stream": True,
          "voice_setting": {"voice_id": "male_0004_a"}},
    stream=True
) as r:
    with open("output.mp3", "wb") as f:
        for line in r.iter_lines():
            if line:
                line_str = line.decode("utf-8")
                if line_str.startswith("data: "):
                    chunk = json.loads(line_str[6:])
                    if chunk.get("data", {}).get("audio"):
                        f.write(bytes.fromhex(chunk["data"]["audio"]))