将流式音频翻译成文本

Media Translation 可将音频文件或语音流翻译为另一种语言的文本。本页面提供的代码示例展示了如何使用 Media Translation 客户端库将流式音频翻译成文本。

设置项目

在使用 Media Translation 之前，您需要先设置一个 Google Cloud 项目，并为该项目启用 Media Translation API。

登录您的 Google Cloud 帐号。如果您是 Google Cloud 新手，请创建一个帐号来评估我们的产品在实际场景中的表现。新客户还可获享 $300 赠金，用于运行、测试和部署工作负载。

在 Google Cloud Console 中的项目选择器页面上，选择或创建一个 Google Cloud 项目。

转到“项目选择器”

确保您的 Google Cloud 项目已启用结算功能。了解如何检查项目是否已启用结算功能。

启用 Media Translation API。

启用 API

创建服务帐号：

在 Google Cloud 控制台中，转到创建服务帐号页面。
转到“创建服务帐号”
选择您的项目。
在服务帐号名称字段中，输入一个名称。Google Cloud 控制台会根据此名称填充服务帐号 ID 字段。

在服务帐号说明字段中，输入说明。例如，Service account for quickstart。
点击创建并继续。
如需提供对项目的访问权限，请向服务帐号授予 Project > Owner 角色。

如需授予该角色，请找到选择角色列表，然后选择 Project > Owner。

注意：角色字段会影响您的服务帐号可以访问项目中的哪些资源。您可以撤消这些角色或稍后授予其他角色。在生产环境中，请勿授予 Owner、Editor 或 Viewer 角色。应授予满足您的需求的预定义角色或自定义角色。
点击继续。
点击完成以完成服务帐号的创建过程。

不要关闭浏览器窗口。您将在下一步骤中用到它。

创建服务帐号密钥：

在 Google Cloud 控制台中，点击您创建的服务帐号的电子邮件地址。
点击密钥。
点击添加密钥，然后点击创建新密钥。
点击创建。JSON 密钥文件将下载到您的计算机上。
点击关闭。

将环境变量 GOOGLE_APPLICATION_CREDENTIALS 设置为包含您的服务帐号密钥的 JSON 文件的路径。此变量仅适用于当前的 shell 会话，因此，如果您打开新的会话，请重新设置该变量。

示例：Linux 或 macOS

export GOOGLE_APPLICATION_CREDENTIALS="KEY_PATH"

将 KEY_PATH 替换为包含您的服务帐号密钥的 JSON 文件的路径。

例如：

export GOOGLE_APPLICATION_CREDENTIALS="/home/user/Downloads/service-account-file.json"

示例：Windows

对于 PowerShell：

$env:GOOGLE_APPLICATION_CREDENTIALS="KEY_PATH"

将 KEY_PATH 替换为包含您的服务帐号密钥的 JSON 文件的路径。

例如：

$env:GOOGLE_APPLICATION_CREDENTIALS="C:\Users\username\Downloads\service-account-file.json"

对于命令提示符：

set GOOGLE_APPLICATION_CREDENTIALS=KEY_PATH

将 KEY_PATH 替换为包含您的服务帐号密钥的 JSON 文件的路径。

安装 Google Cloud CLI。

如需初始化 gcloud CLI，请运行以下命令：

gcloud init

在 Google Cloud Console 中的项目选择器页面上，选择或创建一个 Google Cloud 项目。

转到“项目选择器”

确保您的 Google Cloud 项目已启用结算功能。了解如何检查项目是否已启用结算功能。

启用 Media Translation API。

启用 API

创建服务帐号：

在 Google Cloud 控制台中，转到创建服务帐号页面。
转到“创建服务帐号”
选择您的项目。
在服务帐号名称字段中，输入一个名称。Google Cloud 控制台会根据此名称填充服务帐号 ID 字段。

在服务帐号说明字段中，输入说明。例如，Service account for quickstart。
点击创建并继续。
如需提供对项目的访问权限，请向服务帐号授予 Project > Owner 角色。

如需授予该角色，请找到选择角色列表，然后选择 Project > Owner。

注意：角色字段会影响您的服务帐号可以访问项目中的哪些资源。您可以撤消这些角色或稍后授予其他角色。在生产环境中，请勿授予 Owner、Editor 或 Viewer 角色。应授予满足您的需求的预定义角色或自定义角色。
点击继续。
点击完成以完成服务帐号的创建过程。

不要关闭浏览器窗口。您将在下一步骤中用到它。

创建服务帐号密钥：

在 Google Cloud 控制台中，点击您创建的服务帐号的电子邮件地址。
点击密钥。
点击添加密钥，然后点击创建新密钥。
点击创建。JSON 密钥文件将下载到您的计算机上。
点击关闭。

示例：Linux 或 macOS

export GOOGLE_APPLICATION_CREDENTIALS="KEY_PATH"

将 KEY_PATH 替换为包含您的服务帐号密钥的 JSON 文件的路径。

例如：

export GOOGLE_APPLICATION_CREDENTIALS="/home/user/Downloads/service-account-file.json"

示例：Windows

对于 PowerShell：

$env:GOOGLE_APPLICATION_CREDENTIALS="KEY_PATH"

将 KEY_PATH 替换为包含您的服务帐号密钥的 JSON 文件的路径。

例如：

$env:GOOGLE_APPLICATION_CREDENTIALS="C:\Users\username\Downloads\service-account-file.json"

对于命令提示符：

set GOOGLE_APPLICATION_CREDENTIALS=KEY_PATH

将 KEY_PATH 替换为包含您的服务帐号密钥的 JSON 文件的路径。

安装 Google Cloud CLI。

如需初始化 gcloud CLI，请运行以下命令：

gcloud init

根据您的首选语言安装客户端库。

翻译语音内容

以下代码示例展示了如何翻译通过包含最长五分钟音频的文件或直播麦克风采集到的语音。如需有关如何提供语音数据以取得最佳识别准确率的建议，请参阅最佳做法。

无论音频源如何，主要操作步骤都是相同的：

初始化一个 SpeechTranslationServiceClient 客户端，以用于向 Media Translation 发送请求。

您可以使用同一个客户端重复发出多个请求。
创建一个 StreamingTranslateSpeechConfig 请求对象，以指定如何处理音频。

StreamingTranslateSpeechConfig 对象包含一个 TranslateSpeechConfig 对象（提供有关音频源文件的信息）和一个 single_utterance 属性（指定在讲话人暂停讲话时 Media Translation 是否继续执行翻译）。

TranslateSpeechConfig 对象提供音频源的技术规范（例如其编码和采样率）、设置翻译的源语言和目标语言（使用相应的 BCP-47 语言代码指定），并定义 Media Translation 使用何种翻译模型处理转录。

注意：您可以视需要指定翻译模型。Media Translation 为大多数请求使用基本默认模型。对于某些语言对，我们提供了针对电话通话或视频而优化的增强型模型。如需了解可用模型，请参阅语言支持页面。
发送一系列 StreamingTranslateSpeechRequest 请求对象。

您需要为待翻译的每个音频文件发送一系列请求。第一个请求提供请求的 StreamingTranslateSpeechConfig 对象，随后的请求则以流式传输方式提供音频内容。
接收 StreamingTranslateSpeechResult 响应对象。

虽然会接收 text_translation_result.is_final 值为 false 的任何响应，但最新的翻译结果会覆盖上一个结果。

当 Media Translation 生成最终结果时，text_translation_result.is_final 字段设置为 true，后续收到的任何翻译结果都会附加到上一个结果之后。（在本例中不会覆盖上一个结果）。您可以输出完成的翻译，并从新的部分开始，处理下一部分转录和相应音频。

当讲话人停止时，如果 StreamingTranslateSpeechConfig 请求对象中的 single_utterance 字段设为 true，Media Translation 将为响应中的 speech_event_type 事件返回 END_OF_SINGLE_UTTERANCE 事件。客户端将停止发送请求，但仍会继续接收响应，直到翻译完成。
流式传输的时长上限为 5 分钟。如果超出此上限，系统将返回 OUT_OF_RANGE 错误。

代码示例

翻译音频文件中的语音内容

Java

在 GitHub 上查看


import com.google.api.gax.rpc.BidiStream;
import com.google.cloud.mediatranslation.v1beta1.SpeechTranslationServiceClient;
import com.google.cloud.mediatranslation.v1beta1.StreamingTranslateSpeechConfig;
import com.google.cloud.mediatranslation.v1beta1.StreamingTranslateSpeechRequest;
import com.google.cloud.mediatranslation.v1beta1.StreamingTranslateSpeechResponse;
import com.google.cloud.mediatranslation.v1beta1.StreamingTranslateSpeechResult;
import com.google.cloud.mediatranslation.v1beta1.TranslateSpeechConfig;
import com.google.protobuf.ByteString;
import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.Path;
import java.nio.file.Paths;

public class TranslateFromFile {

  public static void translateFromFile() throws IOException {
    // TODO(developer): Replace these variables before running the sample.
    String filePath = "path/to/audio.raw";
    translateFromFile(filePath);
  }

  public static void translateFromFile(String filePath) throws IOException {
    // Initialize client that will be used to send requests. This client only needs to be created
    // once, and can be reused for multiple requests. After completing all of your requests, call
    // the "close" method on the client to safely clean up any remaining background resources.
    try (SpeechTranslationServiceClient client = SpeechTranslationServiceClient.create()) {
      Path path = Paths.get(filePath);
      byte[] content = Files.readAllBytes(path);
      ByteString audioContent = ByteString.copyFrom(content);

      TranslateSpeechConfig audioConfig =
          TranslateSpeechConfig.newBuilder()
              .setAudioEncoding("linear16")
              .setSampleRateHertz(16000)
              .setSourceLanguageCode("en-US")
              .setTargetLanguageCode("fr-FR")
              .build();

      StreamingTranslateSpeechConfig config =
          StreamingTranslateSpeechConfig.newBuilder()
              .setAudioConfig(audioConfig)
              .setSingleUtterance(true)
              .build();

      BidiStream<StreamingTranslateSpeechRequest, StreamingTranslateSpeechResponse> bidiStream =
          client.streamingTranslateSpeechCallable().call();

      // The first request contains the configuration.
      StreamingTranslateSpeechRequest requestConfig =
          StreamingTranslateSpeechRequest.newBuilder().setStreamingConfig(config).build();

      // The second request contains the audio
      StreamingTranslateSpeechRequest request =
          StreamingTranslateSpeechRequest.newBuilder().setAudioContent(audioContent).build();

      bidiStream.send(requestConfig);
      bidiStream.send(request);

      for (StreamingTranslateSpeechResponse response : bidiStream) {
        // Once the transcription settles, the response contains the
        // is_final result. The other results will be for subsequent portions of
        // the audio.
        StreamingTranslateSpeechResult res = response.getResult();
        String translation = res.getTextTranslationResult().getTranslation();

        if (res.getTextTranslationResult().getIsFinal()) {
          System.out.println(String.format("\nFinal translation: %s", translation));
          break;
        }
        System.out.println(String.format("\nPartial translation: %s", translation));
      }
    }
  }
}

Node.js

在 GitHub 上查看

const fs = require('fs');

// Imports the CLoud Media Translation client library
const {
  SpeechTranslationServiceClient,
} = require('@google-cloud/media-translation');

// Creates a client
const client = new SpeechTranslationServiceClient();

async function translate_from_file() {
  /**
   * TODO(developer): Uncomment the following lines before running the sample.
   */
  // const filename = 'Local path to audio file, e.g. /path/to/audio.raw';
  // const encoding = 'Encoding of the audio file, e.g. LINEAR16';
  // const sourceLanguage = 'BCP-47 source language code, e.g. en-US';
  // const targetLanguage = 'BCP-47 target language code, e.g. es-ES';

  const config = {
    audioConfig: {
      audioEncoding: encoding,
      sourceLanguageCode: sourceLanguage,
      targetLanguageCode: targetLanguage,
    },
    single_utterance: true,
  };

  // First request needs to have only a streaming config, no data.
  const initialRequest = {
    streamingConfig: config,
    audioContent: null,
  };

  const readStream = fs.createReadStream(filename, {
    highWaterMark: 4096,
    encoding: 'base64',
  });

  const chunks = [];
  readStream
    .on('data', chunk => {
      const request = {
        streamingConfig: config,
        audioContent: chunk.toString(),
      };
      chunks.push(request);
    })
    .on('close', () => {
      // Config-only request should be first in stream of requests
      stream.write(initialRequest);
      for (let i = 0; i < chunks.length; i++) {
        stream.write(chunks[i]);
      }
      stream.end();
    });

  const stream = client.streamingTranslateSpeech().on('data', response => {
    const {result} = response;
    if (result.textTranslationResult.isFinal) {
      console.log(
        `\nFinal translation: ${result.textTranslationResult.translation}`
      );
      console.log(`Final recognition result: ${result.recognitionResult}`);
    } else {
      console.log(
        `\nPartial translation: ${result.textTranslationResult.translation}`
      );
      console.log(`Partial recognition result: ${result.recognitionResult}`);
    }
  });

Python

在 GitHub 上查看

from google.cloud import mediatranslation

def translate_from_file(file_path="path/to/your/file"):
    client = mediatranslation.SpeechTranslationServiceClient()

    # The `sample_rate_hertz` field is not required for FLAC and WAV (Linear16)
    # encoded data. Other audio encodings must provide the sampling rate.
    audio_config = mediatranslation.TranslateSpeechConfig(
        audio_encoding="linear16",
        source_language_code="en-US",
        target_language_code="fr-FR",
    )

    streaming_config = mediatranslation.StreamingTranslateSpeechConfig(
        audio_config=audio_config, single_utterance=True
    )

    def request_generator(config, audio_file_path):

        # The first request contains the configuration.
        # Note that audio_content is explicitly set to None.
        yield mediatranslation.StreamingTranslateSpeechRequest(streaming_config=config)

        with open(audio_file_path, "rb") as audio:
            while True:
                chunk = audio.read(4096)
                if not chunk:
                    break
                yield mediatranslation.StreamingTranslateSpeechRequest(
                    audio_content=chunk
                )

    requests = request_generator(streaming_config, file_path)
    responses = client.streaming_translate_speech(requests)

    for response in responses:
        # Once the transcription settles, the response contains the
        # is_final result. The other results will be for subsequent portions of
        # the audio.
        print(f"Response: {response}")
        result = response.result
        translation = result.text_translation_result.translation

        if result.text_translation_result.is_final:
            print("\nFinal translation: {0}".format(translation))
            break

        print("\nPartial translation: {0}".format(translation))

翻译麦克风中的语音内容

Java

在 GitHub 上查看


import com.google.api.gax.rpc.ClientStream;
import com.google.api.gax.rpc.ResponseObserver;
import com.google.api.gax.rpc.StreamController;
import com.google.cloud.mediatranslation.v1beta1.SpeechTranslationServiceClient;
import com.google.cloud.mediatranslation.v1beta1.StreamingTranslateSpeechConfig;
import com.google.cloud.mediatranslation.v1beta1.StreamingTranslateSpeechRequest;
import com.google.cloud.mediatranslation.v1beta1.StreamingTranslateSpeechResponse;
import com.google.cloud.mediatranslation.v1beta1.StreamingTranslateSpeechResult;
import com.google.cloud.mediatranslation.v1beta1.TranslateSpeechConfig;
import com.google.protobuf.ByteString;
import java.io.IOException;
import javax.sound.sampled.AudioFormat;
import javax.sound.sampled.AudioInputStream;
import javax.sound.sampled.AudioSystem;
import javax.sound.sampled.DataLine;
import javax.sound.sampled.LineUnavailableException;
import javax.sound.sampled.TargetDataLine;

public class TranslateFromMic {

  public static void main(String[] args) throws IOException, LineUnavailableException {
    translateFromMic();
  }

  public static void translateFromMic() throws IOException, LineUnavailableException {

    ResponseObserver<StreamingTranslateSpeechResponse> responseObserver = null;

    // Initialize client that will be used to send requests. This client only needs to be created
    // once, and can be reused for multiple requests. After completing all of your requests, call
    // the "close" method on the client to safely clean up any remaining background resources.
    try (SpeechTranslationServiceClient client = SpeechTranslationServiceClient.create()) {
      responseObserver =
          new ResponseObserver<StreamingTranslateSpeechResponse>() {

            @Override
            public void onStart(StreamController controller) {}

            @Override
            public void onResponse(StreamingTranslateSpeechResponse response) {
              StreamingTranslateSpeechResult res = response.getResult();
              String translation = res.getTextTranslationResult().getTranslation();

              if (res.getTextTranslationResult().getIsFinal()) {
                System.out.println(String.format("\nFinal translation: %s", translation));
              } else {
                System.out.println(String.format("\nPartial translation: %s", translation));
              }
            }

            @Override
            public void onComplete() {}

            public void onError(Throwable t) {
              System.out.println(t);
            }
          };

      ClientStream<StreamingTranslateSpeechRequest> clientStream =
          client.streamingTranslateSpeechCallable().splitCall(responseObserver);

      TranslateSpeechConfig audioConfig =
          TranslateSpeechConfig.newBuilder()
              .setAudioEncoding("linear16")
              .setSourceLanguageCode("en-US")
              .setTargetLanguageCode("es-ES")
              .setSampleRateHertz(16000)
              .build();

      StreamingTranslateSpeechConfig streamingRecognitionConfig =
          StreamingTranslateSpeechConfig.newBuilder().setAudioConfig(audioConfig).build();

      StreamingTranslateSpeechRequest request =
          StreamingTranslateSpeechRequest.newBuilder()
              .setStreamingConfig(streamingRecognitionConfig)
              .build(); // The first request in a streaming call has to be a config

      clientStream.send(request);
      // SampleRate:16000Hz, SampleSizeInBits: 16, Number of channels: 1, Signed: true,
      // bigEndian: false
      AudioFormat audioFormat = new AudioFormat(16000, 16, 1, true, false);
      DataLine.Info targetInfo =
          new DataLine.Info(
              TargetDataLine.class,
              audioFormat); // Set the system information to read from the microphone audio stream

      if (!AudioSystem.isLineSupported(targetInfo)) {
        System.out.println("Microphone not supported");
        System.exit(0);
      }
      // Target data line captures the audio stream the microphone produces.
      TargetDataLine targetDataLine = (TargetDataLine) AudioSystem.getLine(targetInfo);
      targetDataLine.open(audioFormat);
      targetDataLine.start();
      System.out.println("Start speaking... Press Ctrl-C to stop");
      long startTime = System.currentTimeMillis();
      // Audio Input Stream
      AudioInputStream audio = new AudioInputStream(targetDataLine);

      while (true) {
        byte[] data = new byte[6400];
        audio.read(data);
        request =
            StreamingTranslateSpeechRequest.newBuilder()
                .setAudioContent(ByteString.copyFrom(data))
                .build();
        clientStream.send(request);
      }
    }
  }
}

Node.js

在 GitHub 上查看


// Allow user input from terminal
const readline = require('readline');

const rl = readline.createInterface({
  input: process.stdin,
  output: process.stdout,
});

function doTranslationLoop() {
  rl.question("Press any key to translate or 'q' to quit: ", answer => {
    if (answer.toLowerCase() === 'q') {
      rl.close();
    } else {
      translateFromMicrophone();
    }
  });
}

// Node-Record-lpcm16
const recorder = require('node-record-lpcm16');

// Imports the Cloud Media Translation client library
const {
  SpeechTranslationServiceClient,
} = require('@google-cloud/media-translation');

// Creates a client
const client = new SpeechTranslationServiceClient();

function translateFromMicrophone() {
  /**
   * TODO(developer): Uncomment the following lines before running the sample.
   */
  //const encoding = 'linear16';
  //const sampleRateHertz = 16000;
  //const sourceLanguage = 'Language to translate from, as BCP-47 locale';
  //const targetLanguage = 'Language to translate to, as BCP-47 locale';
  console.log('Begin speaking ...');

  const config = {
    audioConfig: {
      audioEncoding: encoding,
      sourceLanguageCode: sourceLanguage,
      targetLanguageCode: targetLanguage,
    },
    singleUtterance: true,
  };

  // First request needs to have only a streaming config, no data.
  const initialRequest = {
    streamingConfig: config,
    audioContent: null,
  };

  let currentTranslation = '';
  let currentRecognition = '';
  // Create a recognize stream
  const stream = client
    .streamingTranslateSpeech()
    .on('error', e => {
      if (e.code && e.code === 4) {
        console.log('Streaming translation reached its deadline.');
      } else {
        console.log(e);
      }
    })
    .on('data', response => {
      const {result, speechEventType} = response;
      if (speechEventType === 'END_OF_SINGLE_UTTERANCE') {
        console.log(`\nFinal translation: ${currentTranslation}`);
        console.log(`Final recognition result: ${currentRecognition}`);

        stream.destroy();
        recording.stop();
      } else {
        currentTranslation = result.textTranslationResult.translation;
        currentRecognition = result.recognitionResult;
        console.log(`\nPartial translation: ${currentTranslation}`);
        console.log(`Partial recognition result: ${currentRecognition}`);
      }
    });

  let isFirst = true;
  // Start recording and send microphone input to the Media Translation API
  const recording = recorder.record({
    sampleRateHertz: sampleRateHertz,
    threshold: 0, //silence threshold
    recordProgram: 'rec',
    silence: '5.0', //seconds of silence before ending
  });
  recording
    .stream()
    .on('data', chunk => {
      if (isFirst) {
        stream.write(initialRequest);
        isFirst = false;
      }
      const request = {
        streamingConfig: config,
        audioContent: chunk.toString('base64'),
      };
      if (!stream.destroyed) {
        stream.write(request);
      }
    })
    .on('close', () => {
      doTranslationLoop();
    });
}

doTranslationLoop();

Python

在 GitHub 上查看

from __future__ import division

import itertools

from google.cloud import mediatranslation as media
import pyaudio
from six.moves import queue

# Audio recording parameters
RATE = 16000
CHUNK = int(RATE / 10)  # 100ms
SpeechEventType = media.StreamingTranslateSpeechResponse.SpeechEventType

class MicrophoneStream:
    """Opens a recording stream as a generator yielding the audio chunks."""

    def __init__(self, rate, chunk):
        self._rate = rate
        self._chunk = chunk

        # Create a thread-safe buffer of audio data
        self._buff = queue.Queue()
        self.closed = True

    def __enter__(self):
        self._audio_interface = pyaudio.PyAudio()
        self._audio_stream = self._audio_interface.open(
            format=pyaudio.paInt16,
            channels=1,
            rate=self._rate,
            input=True,
            frames_per_buffer=self._chunk,
            # Run the audio stream asynchronously to fill the buffer object.
            # This is necessary so that the input device's buffer doesn't
            # overflow while the calling thread makes network requests, etc.
            stream_callback=self._fill_buffer,
        )

        self.closed = False

        return self

    def __exit__(self, type=None, value=None, traceback=None):
        self._audio_stream.stop_stream()
        self._audio_stream.close()
        self.closed = True
        # Signal the generator to terminate so that the client's
        # streaming_recognize method will not block the process termination.
        self._buff.put(None)
        self._audio_interface.terminate()

    def _fill_buffer(self, in_data, frame_count, time_info, status_flags):
        """Continuously collect data from the audio stream, into the buffer."""
        self._buff.put(in_data)
        return None, pyaudio.paContinue

    def exit(self):
        self.__exit__()

    def generator(self):
        while not self.closed:
            # Use a blocking get() to ensure there's at least one chunk of
            # data, and stop iteration if the chunk is None, indicating the
            # end of the audio stream.
            chunk = self._buff.get()
            if chunk is None:
                return
            data = [chunk]

            # Now consume whatever other data's still buffered.
            while True:
                try:
                    chunk = self._buff.get(block=False)
                    if chunk is None:
                        return
                    data.append(chunk)
                except queue.Empty:
                    break

            yield b"".join(data)

def listen_print_loop(responses):
    """Iterates through server responses and prints them.

    The responses passed is a generator that will block until a response
    is provided by the server.
    """
    translation = ""
    for response in responses:
        # Once the transcription settles, the response contains the
        # END_OF_SINGLE_UTTERANCE event.
        if response.speech_event_type == SpeechEventType.END_OF_SINGLE_UTTERANCE:

            print("\nFinal translation: {0}".format(translation))
            return 0

        result = response.result
        translation = result.text_translation_result.translation

        print("\nPartial translation: {0}".format(translation))

def do_translation_loop():
    print("Begin speaking...")

    client = media.SpeechTranslationServiceClient()

    speech_config = media.TranslateSpeechConfig(
        audio_encoding="linear16",
        source_language_code="en-US",
        target_language_code="es-ES",
    )

    config = media.StreamingTranslateSpeechConfig(
        audio_config=speech_config, single_utterance=True
    )

    # The first request contains the configuration.
    # Note that audio_content is explicitly set to None.
    first_request = media.StreamingTranslateSpeechRequest(streaming_config=config)

    with MicrophoneStream(RATE, CHUNK) as stream:
        audio_generator = stream.generator()
        mic_requests = (
            media.StreamingTranslateSpeechRequest(audio_content=content)
            for content in audio_generator
        )

        requests = itertools.chain(iter([first_request]), mic_requests)

        responses = client.streaming_translate_speech(requests)

        # Print the translation responses as they arrive
        result = listen_print_loop(responses)
        if result == 0:
            stream.exit()

def main():
    while True:
        print()
        option = input("Press any key to translate or 'q' to quit: ")

        if option.lower() == "q":
            break

        do_translation_loop()

if __name__ == "__main__":
    main()