感受好久没写中文技术文章了。说实话,学东西都是基于英文,或者 别人从英文翻译成中文 咱们再捡二手货
学习。因此用中文写技术文章怎么都感受是在骗人
,怎么都以为很别扭
。编程
可是这一次的主角是百度。框架
虽然认真来说,全部编程语言、框架、核心技术都是外国人写的(开源
),但彷佛你拼凑一下,仍能够贴上国产的标签(这个就至关于零件不是我作的,但我用它拼出了一件产品,产品是个人)。编程语言
并且又加上这个 API 是免费的,因此我能够介绍一下。(最近几年百度为了拿钱换名气
,在海外仍是作了很多工做,好比创建了 Twitter、Github 帐户,成立了 AI研究室,开源了一些项目)学习
我这我的不喜欢讲废话:翻译
# Author: yingshaoxo #### For baidu voice from aip import AipSpeech APP_ID = '15311704' API_KEY = 'yTzBl40WBlhFOo1GnKk0YQTN' SECRET_KEY = 'xpWedO1u0ZLATHijhetFo7dE5ibMsI6Q' client = AipSpeech(APP_ID, API_KEY, SECRET_KEY) def get_text(wav_bytes): result = client.asr(wav_bytes, 'wav', 16000, {'dev_pid': 1536,}) try: text = result['result'][0] except Exception as e: print(e) text = "" return text #### For real time voice recording import speech_recognition as sr r = sr.Recognizer() mic = sr.Microphone() while 1: print("\nPlease try to speak something...") with mic as source: r.adjust_for_ambient_noise(source) audio = r.listen(source) audio_data = audio.get_wav_data(convert_rate=16000) print("\nGot you, now I'm trying to recognize that...") text = get_text(audio_data) print(f"\n{text}")