Create Cute Voiceovers with Baidu TTS and Python

During the National Day holiday many elders use voice features in news apps; text‑to‑speech can help them read content without straining their eyes.

Install Baidu SDK

Install the Baidu AI Python SDK via pip, using a faster mirror.

pip3 install baidu-aip -i https://pypi.douban.com/simple/

Create Application

Visit the Baidu TTS page ( https://ai.baidu.com/tech/speech/tts ), log in with a Baidu account, and create a new application. Select only the speech synthesis API if you only need text‑to‑speech. After creation you will obtain three essential credentials: AppID, API Key, and Secret Key.

The application list shows the three keys needed for API calls.

In the documentation you can see the basic request parameters.

Code Debugging

Write a simple script that calls the Baidu TTS API. Fill in the AppID, API Key, and Secret Key you obtained.

from aip import AipSpeech
APP_ID = '输入你的APP_ID'
API_KEY = '输入你的API_KEY'
SECRET_KEY = '输入你的SECRET_KEY'
client = AipSpeech(APP_ID, API_KEY, SECRET_KEY)
# 中文：zh 粤语：ct 英文：en
result = client.synthesis('哈哈哈哈', 'zh', 1, {
    'vol': 5, 'per': 4
})
if not isinstance(result, dict):
    with open('audio.mp3', 'wb') as f:
        f.write(result)

Running the script generates an audio.mp3 file in the current directory.

Add Daily Quotes for a Cute Voice

Use the Jinshan Ciba daily sentence API to fetch an English sentence and its Chinese translation, then let the TTS engine read them, giving the voice a more playful tone.

def get_msg():
    url = 'http://open.iciba.com/dsapi/'  # Jinshan Ciba daily sentence API
    html = requests.get(url)
    content = html.json()['content']  # English sentence
    note = html.json()['note']        # Chinese translation
    return content, note

Replace the fixed text in the previous example with the sentences returned by get_msg(), then run the script to produce a cute voiceover.