跳转到主要内容
Send a POST request to https://api.qhaigc.net/v1/audio/speech to convert a text string into an audio file. The qhai-tts model supports over 38 distinct voice characters — from cute and youthful to mature and authoritative — including popular game and anime personas such as 派蒙 (Paimon), 温迪 (Venti), and 纳西妲 (Nahida).
Call GET /v1/audio/voices to retrieve the full list of available voice character names before making a TTS request.

​
Request

POST https://api.qhaigc.net/v1/audio/speech

​
Parameters

​
model
string
必填
The TTS model to use. Set this to "qhai-tts".
​
input
string
必填
The text to convert to speech. Maximum length is approximately 4096 characters.
​
voice
string
必填
The name of the voice character to use, for example "可爱少女" or "御姐". Call GET /v1/audio/voices to see all available characters.
​
response_format
string
The audio format of the returned file. Supported values include "mp3", "wav", and "opus". Defaults to "mp3".
​
speed
number
Playback speed multiplier. A value of 1.0 is normal speed. Increase to speed up or decrease to slow down the output. Defaults to 1.0.

​
Response

The endpoint returns the audio file as a binary stream in the format specified by response_format. Save the response body directly to a file.

​
Code examples

import openai

client = openai.OpenAI(
    api_key="sk-your-api-key-here",
    base_url="https://api.qhaigc.net/v1"
)

response = client.audio.speech.create(
    model="qhai-tts",
    voice="可爱少女",
    input="启航 AI API 系统是一款聚焦低成本探索 AIGC 无限可能的服务平台。"
)

response.stream_to_file("output.mp3")