Build a Python Chatbot with Turing API: Step‑by‑Step Guide

Introduction

While at home I built a small project: a voice‑chat robot based on Python and shared it. The overall project is simple, the official documentation is very detailed, and you can get started quickly.

Goal

Deploy a Turing robot on the desktop to serve as a work assistant or companion for casual conversation.

Libraries Used

Version 1.0: requests, json Version 2.0: requests, json, selenium (to automatically open URLs returned by the Turing API).

Implementation

1. Create a Turing robot – the creation process is straightforward and unrelated to Python, so it is omitted here. After creation, you must authenticate to enjoy 100 free replies per day.

2. Write the code (Version 1.0)

import requests</code><code>import json</code><code>from selenium import webdriver

Set the API endpoint:

url = 'http://openapi.tuling123.com/openapi/api/v2'

Define request headers:

headers = {'User-Agent':'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36'}

Enter a loop to read user input and send it to the API:

while True:</code><code>    aa = input('我：')</code><code>    data = {</code><code>        "perception": {</code><code>            "inputText": {"text": aa},</code><code>            "selfInfo": {"location": {"city": "济南"}}</code><code>        },</code><code>        "userInfo": {"apiKey": "YOUR_API_KEY", "userId": "ANY_ID"}</code><code>    }</code><code>    res = requests.post(url, headers=headers, data=json.dumps(data))</code><code>    try:</code><code>        print('机器人:' + res.json()['results'][0]['values']['text'])</code><code>        if aa == '退出':</code><code>            break</code><code>    except:</code><code>        print('机器人:' + res.json()['results'][0]['values']['url'])

The loop runs indefinitely; typing “退出” exits it. If the API returns a URL instead of text, the program opens the URL.

Version 2.0 Enhancements

Install the selenium library and download ChromeDriver. Place the chromedriver.exe in the Python installation directory.

Add three lines to the previous code to handle URL responses:

except:</code><code>    print('机器人:' + res.json()['results'][0]['values']['url'])</code><code>    url_1 = res.json()['results'][0]['values']['url']</code><code>    driver = webdriver.Chrome()</code><code>    driver.get(url_1)

This automatically opens the returned URL in a browser.

Packaging

Finally, package the script into an executable with: pyinstaller -F path/to/your_script.py The resulting .exe can be placed on the desktop or shared with others.

Conclusion

1. The project is straightforward and the official documentation enables rapid development.

2. webdriver is very useful for bypassing anti‑scraping measures on many sites.

3. The Turing API offers many more features to explore.

4. This chatbot demonstrates how ordinary users can experience AI conveniently.