创造动态艺术：AI在视觉和声音上的突破

A middle-aged man, standing frontally, with a calm and serene expression, embodying wisdom and maturity. His hair is slightly graying, showing signs of aging yet well-maintained. His facial features are moderate, with deep and introspective eyes. He is dressed in a simple yet elegant manner, wearing a well-fitted dark shirt and neat trousers. The background is plain and unobtrusive, highlighting the subject's demeanor and expression. The overall image conveys a sense of steadiness and wisdom that comes with experience, front view, --ar 16:9

from pathlib import Path
from openai import OpenAI

client = OpenAI(
    api_key="sk-XXX"
)
text = """
有钱人以后再也不会向下社交了

我发现我们的社会正在逐渐分裂为不同的层级，这种分裂在很大程度上是由网络上的攻击性评论和偏见造成的。有钱人越来越少地与不同社会阶层的人交流。我尝试邀请一些朋友参加我的节目，他们都婉拒了，担心公开发声会被误解为炫耀，甚至遭受网络攻击。这让我意识到，我们正在失去一种宝贵的东西：不同声音的交流和理解。

更重要的是，这种社会的分层化导致了信息的筛选。我们越来越多地只听到自己想听的声音，这加剧了阶层间的隔离。难道我们就愿意生活在一个只有单一声音的世界里吗？

我们需要跨越社会层级的界限，努力去理解和接受不同的观点和生活方式。我们需要的不是键盘上的攻击，而是面对面的沟通和理解。

让我们一起努力创造一个更加开放、包容的社会。记住，每个人的声音都值得被听到。让我们一起为一个更好的明天努力吧。

"""
speech_file_path = Path(__file__).parent / "speech.mp3"
print(speech_file_path)

response = client.audio.speech.create(
    model="tts-1-hd",
    voice="onyx",
    input=text,
    speed=1.3
)

response.stream_to_file(speech_file_path)