Skip to main content

A simple and easy-to-use library that connects to common artificial intelligence interfaces, allowing for quick development of local web applications. It includes mainstream AI capabilities such as LLM text interaction, image understanding, audio understanding, image generation, speech recognition, and speech synthesis.

Project description

aiZero是一个简单易用的可以连接常用人工智能接口,快速搭建可视化本地web应用的Python第三方库,适配幻码人工智能初阶课程。

快速开始

from aiZero import AIWebApp

# 设定web应用的功能
def my_ai_function():
    pass

app = AIWebApp(title='人工智能助手')    # 初始化web应用
app.set_apikey('YOUR_API_KEY')    # 设定AI接口的api key
app.add_input_text()    # 在页面中添加一个输入文本框
app.add_submit(my_ai_function)    # 添加一个提交按钮,点击后执行函数
app.run(port=6060)    # 启动应用,设定端口,默认为5000

启动后,你将可以在浏览器中访问http://127.0.0.1:6060/ 查看搭建好的web应用。

如果需要实现AI功能的可视化呈现,你只需要将'YOUR_API_KEY'替换为真实的api key,并完善my_ai_function函数即可。

AI功能实现例子

大模型文字交互

单轮对话
from aiZero import AIWebApp, text_generation

def my_ai_function():
    text = app.input_text    # 获取输入文本框的文字内容
    reply = text_generation([text])    # 调用AI接口,获取回复反馈
    app.results.append({'text': reply})    # 以文字形式将结果推送至前端

app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_input_text()
app.add_submit(my_ai_function)
app.run(port=6060)
多轮对话
from aiZero import AIWebApp, text_generation

def my_ai_function():
    text = app.input_text
    chat_history.append(text)    # 将输入内容添加到对话历史
    reply = text_generation(chat_history)
    chat_history.append(reply)    # 将回复结果添加到对话历史
    app.results.append({'text': reply})

chat_history = []    # 储存对话历史
app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_input_text()
app.add_submit(my_ai_function)
app.run(port=6060)
设置系统指令
from aiZero import AIWebApp, text_generation

def my_ai_function():
    text = app.input_text
    chat_history.append(text)
    # prompt参数可以设定系统指令,设定模型的行为要求
    reply = text_generation(chat_history, prompt='你是一个10岁的小学生,名叫小幻')
    chat_history.append(reply)
    app.results.append({'text': reply})

chat_history = []
app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_input_text()
app.add_submit(my_ai_function)
app.run(port=6060)

图像理解

上传图片文件
from aiZero import AIWebApp, image_understanding

def my_ai_function():
    text = app.input_text
    img = app.input_pic    # 获取上传的图片
    reply = image_understanding(img, [text])
    app.results.append({'text': reply})

chat_history = []
app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_input_text()
app.add_pic_file()    # 添加一个图片文件上传按钮
app.add_submit(my_ai_function)
app.run(port=6060)
摄像头捕获图像
from aiZero import AIWebApp, image_understanding

def my_ai_function():
    text = app.input_text
    img = app.input_pic    # 获取捕获的图片
    reply = image_understanding(img, [text])
    app.results.append({'text': reply})

chat_history = []
app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_input_text()
app.add_camera()    # 添加摄像头捕获窗口和按钮
app.add_submit(my_ai_function)
app.run(port=6060)

声音理解

上传音频文件
from aiZero import AIWebApp, audio_understanding

def my_ai_function():
    text = app.input_text
    audio = app.input_audio    # 获取上传的音频
    reply = audio_understanding(audio, [text])
    app.results.append({'text': reply})

chat_history = []
app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_input_text()
app.add_audio_file()    # 添加一个音频文件上传按钮
app.add_submit(my_ai_function)
app.run(port=6060)
实时录音
from aiZero import AIWebApp, audio_understanding

def my_ai_function():
    text = app.input_text
    audio = app.input_audio    # 获取录制的音频
    reply = audio_understanding(audio, [text])
    app.results.append({'text': reply})

chat_history = []
app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_input_text()
app.add_record()    # 添加音频录制功能区
app.add_submit(my_ai_function)
app.run(port=6060)

图像生成

from aiZero import AIWebApp, image_generation

def my_ai_function():
    text = app.input_text
    reply = image_generation(text)
    app.results.append({'image': reply})    # 将生成的图像推送到前端

app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_input_text()
app.add_submit(my_ai_function)
app.run(port=6060)

人物图像风格重绘

from aiZero import AIWebApp, human_repaint

def my_ai_function():
    img = app.input_pic
    reply = human_repaint(img)
    app.results.append({'image': reply})

app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_camera()
app.add_submit(my_ai_function)
app.run(port=6060)

以上例程为使用摄像头采集图像,也可以改为上传图片文件。

human_repaint函数可以接受style参数设定风格类型,可选值为0~9的数字(默认值为7)。

涂鸦作画

from aiZero import AIWebApp, sketch_to_image

def my_ai_function():
    text = app.input_text    # 涂鸦作画的提示文字
    img = app.input_pic    # 涂鸦草图图像
    reply = sketch_to_image(img, text)
    app.results.append({'image': reply})

app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_input_text()
app.add_pic_file()
app.add_submit(my_ai_function)
app.run(port=6060)

以上例程为上传图片文件,也可以改为用摄像头捕获图像。

sktech_to_image函数可以接受style参数设定风格类型,包括:

  • "<3d cartoon>":3D 卡通
  • "<anime>":二次元(默认值)
  • "<oil painting>":油画
  • "<watercolor>" :水彩
  • "<flat illustration>":扁平插画

语音识别

from aiZero import AIWebApp, speech_recognition

def my_ai_function():
    audio = app.input_audio
    reply = speech_recognition(audio)
    app.results.append({'text': reply})

app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_record()
app.add_submit(my_ai_function)
app.run(port=6060)

所用接口支持中英文双语的语音识别。以上例程为实时录音,也可以改为上传录音音频文件。

语音合成

from aiZero import AIWebApp, speech_synthesis

def my_ai_function():
    text = app.input_text
    reply = speech_synthesis(text)
    app.results.append({'audio': reply})    # 将音频结果推送到前端

app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_input_text()
app.add_submit(my_ai_function)
app.run(port=6060)

speech_synthesis函数可以接受以下参数:

  • model:设定使用的语音模型,详细列表参见链接
  • rate:设定语速快慢,取值范围0.5~2,默认值为1。
  • pitch:设定语调高低,取值范围0.5~2,默认值为1。

进阶使用

多种AI功能的混合

aiZero支持方便地将不同的AI功能进行链式整合,例如,将文字交互、语音识别、语音合成结合起来,可以实现语音交互。

from aiZero import AIWebApp, text_generation, speech_recognition, speech_synthesis

def my_ai_function():
    audio = app.input_audio
    text = speech_recognition(audio)
    chat_history.append(text)
    reply = text_generation(chat_history)
    chat_history.append(reply)
    reply_audio = speech_synthesis(reply)
    app.results.append({'audio': reply_audio})

chat_history = []
app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_record()
app.add_submit(my_ai_function)
app.run(port=6060)

同时输出多种不同类型的信息

aiZero支持同时输出文字、音频和图像等多种不同类型的信息,但每种类型的信息只能为一个。例如,我们可以输入文字,同时以文本和音频形式输出AI的响应。

from aiZero import AIWebApp, text_generation, speech_synthesis

def my_ai_function():
    text = app.input_text
    chat_history.append(text)
    reply = text_generation(chat_history)
    chat_history.append(reply)
    reply_audio = speech_synthesis(reply)
    app.results.append({'text': reply, 'audio': reply_audio})

chat_history = []
app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_input_text()
app.add_submit(my_ai_function)
app.run(port=6060)

同次分时输出多个信息

aiZero也支持在一次大模型交互的输出时,分不同时间输出不同信息。例如,我们制作一个绘图助手,它可以将用户输入的简易文字信息扩充为详细的画面信息后,再交给图像生成功能完成图像的创作。在这个功能设计中,AI的文字回复响应很快而图像生成响应较慢,因此应当在不同时间推送至前端。

from aiZero import AIWebApp, text_generation, image_generation

def my_ai_function():
    text = app.input_text
    chat_history.append(text)
    reply = text_generation(chat_history, prompt='你是一个人工智能绘图助手,你负责将用户输入信息中描述的画面扩充为详细的绘图AI的提示词')
    chat_history.append(reply)
    # 结果推送时包含'running'键,则会继续监听输出结果
    app.results.append({'text': reply, 'running': True})
    img = image_generation(reply)
    # 结果推送时不包含'running'键,则输出结束
    app.results.append({'image': img})

chat_history = []
app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_input_text()
app.add_submit(my_ai_function)
app.run(port=6060)

为了能让前端在收到第一次的文本输出后不会结束对其他输出的监听,必须在输出未完全结束时,在结果中添加'running'键,而在输出完全结束时,推送不包含'running'键的结果。

文本流式输出

在大模型文字交互(text_generation)、图像理解(image_understanding)、声音理解(audio_understanding)等功能中,大模型回复的文本信息支持以流式形式输出,即逐步输出文本内容而非一次性输出全部文本。设置方法只需要将函数的可选参数stream设为True即可,以基础的文字交互为例:

from aiZero import AIWebApp, text_generation

def my_ai_function():
    text = app.input_text
    chat_history.append(text)
    # 设置stream参数值为True,返回值是一个生成器,可用for循环迭代地输出结果
    reply = text_generation(chat_history, stream=True)
    for r in reply:
        # 流式输出过程中,结果推送时须包含'running'键
        app.results.append({'text': r, 'running': True})
    chat_history.append(r)
    # 最后,必须再次推送一个不包含'running'键的结果,才能结束输出
    app.results.append({'text': r})

chat_history = []
app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_input_text()
app.add_submit(my_ai_function)
app.run(port=6060)

错误信息的捕获与处理

aiZero中,所有错误信息都会以'错误:'作为开始部分输出,因此,可以在获取输出信息时,先判断是否为错误信息,再进行相应的推送。

from aiZero import AIWebApp, text_generation

def my_ai_function():
    text = app.input_text
    chat_history.append(text)
    reply = text_generation(chat_history)
    if reply.startswith('错误:'):
        app.results.append({'text': reply})
        chat_history.pop()
    else:
        chat_history.append(reply)
        app.results.append({'text': reply})

chat_history = []
app = AIWebApp(title='人工智能助手')
app.set_apikey('YOUR_API_KEY')
app.add_input_text()
app.add_submit(my_ai_function)
app.run(port=6060)

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

aiZero-0.0.3a1.tar.gz (33.6 kB view details)

Uploaded Source

Built Distribution

aiZero-0.0.3a1-py3-none-any.whl (247.9 kB view details)

Uploaded Python 3

File details

Details for the file aiZero-0.0.3a1.tar.gz.

File metadata

  • Download URL: aiZero-0.0.3a1.tar.gz
  • Upload date:
  • Size: 33.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.0.0 CPython/3.10.3

File hashes

Hashes for aiZero-0.0.3a1.tar.gz
Algorithm Hash digest
SHA256 7518b79322ba44839306be760fe78df5ef220fbdea5af339020f71074ee240ce
MD5 b552c1c856f9dce201e0fa70a3b8532f
BLAKE2b-256 68c1a82602e55f203aa4e6b9c794c473d60e91b8f5574c3d6a2f4ed0139ac35e

See more details on using hashes here.

File details

Details for the file aiZero-0.0.3a1-py3-none-any.whl.

File metadata

  • Download URL: aiZero-0.0.3a1-py3-none-any.whl
  • Upload date:
  • Size: 247.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.0.0 CPython/3.10.3

File hashes

Hashes for aiZero-0.0.3a1-py3-none-any.whl
Algorithm Hash digest
SHA256 30e1707ecc4152c654e90450b66d27530de25d338e8860ed5b0aa27d3412adf9
MD5 ff5d69eb663ffb3ad61267f1c81434fe
BLAKE2b-256 b7b98ee4a598cafbf8669d67bc5dac16a8255aa119ecf0d3a91787b76167be6f

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page