使用 Python 创建实时语音翻译器

17 Mar 2025 | 5 分钟阅读

在本教程中，我们将使用 Python 开发一个实时语音翻译器。

所需模块

以下是创建实时翻译器所需的库或模块：

playsound：此模块用于播放 Python 中的声音。

Googletrans：“Googletrans”模块可免费使用，并且是一个无限制的 Python 库，实现了 Google Translate API。

语音识别模块：这是一个 Python 可以用来识别所提供命令的库。pip 是执行语音识别的首选方法。

PyAudio：此模块用于接收用户的命令。

gTTs：gTTs API 支持多种语言，例如英语、印地语、泰米尔语、法语、德语等。

!pip3 install gTTs
!pip3 install gTTS-token

基本思想

一个实时翻译器，可以将人类语音输入翻译成所需的输出语音。它使用 Google 的googleTrans API 和 Python 的speech_recognition 库构建。该程序会将文本从一种语言转换为另一种语言，并将其保存为mp3 格式的音频文件。playsound 模块将用于播放创建的mp3 格式文件。之后，使用 OS 模块删除创建的 MP3 文件。

分步实现

第 1 步：导入所有必需的模块。

from playsound import playsound as PS
import speech_recognition as s_r 
from googletrans import Translator as Trans 
from gtts import gTTS 
import os
import pyaudio

第 2 步：创建所有语言及其代码的元组。

Create a Real Time Voice Translator using Python

第 3 步：从用户端接收语音命令。

# First, we will capture the user's Voice command
# It will take the command through built-in microphone of the device
def take_command():
    r1 = s_r.Recognizer()
    with s_r.Microphone() as source:
        print ("listening to the voice...")
        r1.pause_threshold = 1
        audio1 = r1.listen(source)
  
    try:
        print ("Recognizing the voice...")
        query_1 = r1.recognize_google(audio, language = 'en-in')
        print (f" user said? {query_1}\n")
    except Exception as ep:
# just in case it didn't recognise the voice or language properly
        print ("The user is requested to please say that again...")
        return "None"
    return query_1

第 4 步：从用户端接收语音输入。

# Here, we will take the user's voice input from the user end
query_1 = take_command()
while (query_1 == "None"):
    query_1 = take_command()

第 5 步：从用户端实现目标语言的输入，并将用户输入与语言代码进行映射。

def destination_language():
    print("Please enter the language in which you want to convert the above input \
    : Ex. English, Hindi, German, French etc.")
    print()
  
    # Now, we will implement the input destination language in which the user
# wants to translate the voice command:
    to_language = take_command()
    while (to_language == "None"):
        to_language = take_command()
    to_language = to_language.lower()
    return to_language
  
to_language = destination_language()
  
# Now, we will map the input destination language with the code
while (to_language not in dic_language):
    print ("The language in which the user wants to convertthe voice command\ is currently not available, the user is requested to input some other language")
    print()
    to_language = destination_language()
  
to_language = dic_language[dic_language.index(to_language)+1]

第 6 步：现在我们将调用翻译器。

# Here, we will invoke the Google Translator
translator1 = Trans()

第 7 步：现在，我们将从源语言翻译到目标语言。

# Now we will translate from src to dest
text_to_translate_1 = translator1.translate1(query_1, dest = to_language)
text1 = text_to_translate_1.text

第 8 步：现在，保存翻译后的文件并在播放后删除它们。

# We will be using the Google-Text-to-Speech i.e., gTTS() function of the gtts
# module for speaking the translated text into the input destination language
# selected by the user which is stored in to_language.
# We have also given third argument as False because it speaks very slowly by # default
speak = gTTS(text = text1, language = to_language, slow = False)
  
# We will be using the save() function for saving the translated speech in #captured_JTP_voice.mp3 file
speak.save("captured_JTP_voice.mp3")
  
# Now at last, we will be using the OS module for running the translated voice.
PS('captured_JTP_voice.mp3')
os.remove('captured_JTP_voice.mp3')
print(text)

完整代码实现

from playsound import playsound as PS
import speech_recognition as s_r 
from googletrans import Translator as Trans 
from gtts import gTTS 
import os
import pyaudio

# First, we will capture the user's Voice command
# It will take the command through built-in microphone of the device
def take_command():
    r1 = s_r.Recognizer()
    with s_r.Microphone() as source:
        print ("listening to the voice...")
        r1.pause_threshold = 1
        audio1 = r1.listen(source)
  
    try:
        print ("Recognizing the voice...")
        query_1 = r1.recognize_google(audio, language = 'en-in')
        print (f" user said? {query_1}\n")
    except Exception as ep:
# just in case it didn't recognise the voice or language properly
        print ("The user is requested to please say that again...")
        return "None"
    return query_1
# Here, we will take the user's voice input from the user end
query_1 = take_command()
while (query_1 == "None"):
    query_1 = take_command()

def destination_language():
    print("Please enter the language in which you want to convert the above input \
    : Ex. English, Hindi, German, French etc.")
    print()
  
    # Now, we will implement the input destination language in which the user
# wants to translate the voice command:
    to_language = take_command()
    while (to_language == "None"):
        to_language = take_command()
    to_language = to_language.lower()
    return to_language
  
to_language = destination_language()
  
# Now, we will map the input destination language with the code
while (to_language not in dic_language):
    print ("The language in which the user wants to convertthe voice command\ is currently not available, the user is requested to input some other language")
    print()
    to_language = destination_language()
  
to_language = dic_language[dic_language.index(to_language)+1]
# Here, we will invoke the Google Translator
translator1 = Trans()
# Now we will translate from src to dest
text_to_translate_1 = translator1.translate1(query_1, dest = to_language)
text1 = text_to_translate_1.text
# We will be using the Google-Text-to-Speech i.e., gTTS() function of the gtts
# module for speaking the translated text into the input destination language
# selected by the user which is stored in to_language.
# We have also given third argument as False because it speaks very slowly by # default
speak = gTTS(text = text1, language = to_language, slow = False)
  
# We will be using the save() function for saving the translated speech in #captured_JTP_voice.mp3 file
speak.save("captured_JTP_voice.mp3")
  
# Now at last, we will be using the OS module for running the translated voice.
PS('captured_JTP_voice.mp3')
os.remove('captured_JTP_voice.mp3')
print(text)

输出

下一主题如何在 Python 中对元组进行排序

使用 Python 创建实时语音翻译器

所需模块

分步实现

完整代码实现

联系信息

关注我们

教程

面试题

在线编译器

Python

Java

.Net Framework

AI, ML and Data Science

Cloud Technology

B.Tech and MCA

Web Technology

PHP

Software Testing

Technical Interview

Java Interview

Python

Web Interview

Database Interview

B.Tech / MCA

Important Interview

Software Testing Interview

Company Interviews

Online Compilers

Multiple Choice Questions

Python 问题

使用 Python 创建实时语音翻译器

所需模块

分步实现

完整代码实现

相关帖子

如何在 Python 中绘制曼德勃罗集

Python 中的物理计算

使用 Python 构建 Telegram 机器人

Sklearn predict 函数

Python %s - 字符串格式化

Django vs. Node JS

Python 中的 XGBoost ML 模型

Python 字符串操作

PyScript 教程 | 在 Web 浏览器中运行 Python 脚本

Python 中的 Xticks

订阅 Tpoint Tech

联系信息

关注我们

教程

面试题

在线编译器