前言
现在语音助手已经成为我们生活中ai的代表，下面让我们看一下语音助手的语音输入功能
准备
让我们搜索一下百度的开放平台
进入百度开放平台后选择短语音识别
然后去领取免费的资源
创建应用
记住相关的key
下载库
本次用到的库（以及引入方法）：
import speech_recognition as sr from aip import AipSpeech import pyttsx3 123
代码实现以及
编写我的key
APP_ID = '27468776' API_KEY = 'ygQgtR4DIh14lDi8gwSMCm7o' SECRET_KEY = '7flGlZba42ZYxHudy3dSQ8YW4n89W5hc' client = AipSpeech(APP_ID, API_KEY, SECRET_KEY) 1234
语音的录入：
try: def rec(rate=16000): r = sr.Recognizer() with sr.Microphone(sample_rate=rate) as source: print("请说话") audio = r.listen(source) with open("recording.wav", "wb") as f: f.write(audio.get_wav_data()) except: print("语音录入发生错误") 1234567891011'
使用百度语音作为STT引擎并通过pyttsx3函数将返回的内容读出来形成简单的对话
with open('recording.wav', 'rb') as f: audio_data = f.read() result = client.asr(audio_data, 'wav', 16000, { 'dev_pid': 1536, }) try: result_text = result["result"][0] print("你说: " + result_text) if result_text =="于金龙": engine = pyttsx3.init() engine.say('真帅') engine.runAndWait() elif result_text=="张新凯": engine = pyttsx3.init() engine.say('真难看') engine.runAndWait() except: print("没有识别到语音")
123456789101112131415161718192021完整代码
import pyttsx3 import speech_recognition as sr from aip import AipSpeech # Baidu Speech API, replace with your personal key APP_ID = '27468776' API_KEY = 'ygQgtR4DIh14lDi8gwSMCm7o' SECRET_KEY = '7flGlZba42ZYxHudy3dSQ8YW4n89W5hc' client = AipSpeech(APP_ID, API_KEY, SECRET_KEY) # Use SpeechRecognition to record try: def rec(rate=16000): r = sr.Recognizer() with sr.Microphone(sample_rate=rate) as source: print("请说话") audio = r.listen(source) with open("recording.wav", "wb") as f: f.write(audio.get_wav_data()) except: print("语音录入发生错误") # 使用百度语音作为STT引擎 def listen(): with open('recording.wav', 'rb') as f: audio_data = f.read() result = client.asr(audio_data, 'wav', 16000, { 'dev_pid': 1536, }) try: result_text = result["result"][0] print("你说: " + result_text) if result_text =="阿龙": engine = pyttsx3.init() engine.say('真帅') engine.runAndWait() elif result_text=="其他人": engine = pyttsx3.init() engine.say('真难看') engine.runAndWait() except: print("没有识别到语音") while True: rec() request = listen()
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354运行结果
