Java 如何让斯芬克斯只识别单词

我正在使用实时语音识别器 它识别一些单词，但我只需要这些世界作为输出，但控制台显示 请告诉我如何避免它

  public class Indian {

  public static void main(String[] args) throws Exception {
    Configuration configuration = new Configuration();
    configuration.setAcousticModelPath("resource:/edu/cmu/sphinx/models/en-us/en-us");
    configuration.setDictionaryPath("resource:/edu/cmu/sphinx/models/en-us/cmudict-en-us.dict");
    configuration.setLanguageModelPath("resource:/edu/cmu/sphinx/models/en-us/en-us.lm.bin");
    configuration.setUseGrammar(false);

     LiveSpeechRecognizer recognizer = new     LiveSpeechRecognizer(configuration);

    recognizer.startRecognition(true);
    System.out.println("Start speaking. \n");
    SpeechResult result;
     long timer=0;

 while (timer < 2*60*1000) {
result = recognizer.getResult();
if (result != null) {
           for(WordResult word : result.getWords()) {
            System.out.println("You said: " + word + '\n');
        }
}
  else {
            System.out.println("I can't hear what you said.\n");
        }
 }
 recognizer.stopRecognition();
      }
        }

---

您只需要重新格式化输出。现在，您正在每行打印它们，并使用类

WordResult

的通用

toString

请尝试以下操作：

public class Indian
{
    public static void main(String[] args) throws Exception
    {
        Configuration configuration = new Configuration();
        configuration.setAcousticModelPath("resource:/edu/cmu/sphinx/models/en-us/en-us");
        configuration.setDictionaryPath("resource:/edu/cmu/sphinx/models/en-us/cmudict-en-us.dict");
        configuration.setLanguageModelPath("resource:/edu/cmu/sphinx/models/en-us/en-us.lm.bin");
        configuration.setUseGrammar(false);

        LiveSpeechRecognizer recognizer = new LiveSpeechRecognizer(configuration);

        recognizer.startRecognition(true);
        System.out.println("Start speaking. \n");
        SpeechResult result;
        long timer = 0;

        while (timer < 2 * 60 * 1000)
        {
            result = recognizer.getResult();
            if (result != null)
            {
                System.out.print("You said: ");
                for (WordResult word : result.getWords())
                {
                    if (!word.isFiller())
                        System.out.print(word.getWord() + " ");
                }
                System.out.println();
            }
            else
            {
                System.out.println("I can't hear what you said.\n");
            }
        }
        recognizer.stopRecognition();
    }
}

public-class-Indian
{
公共静态void main（字符串[]args）引发异常
{
配置=新配置（）；
setAcousticModelPath（“资源：/edu/cmu/sphinx/models/en-us/en-us”）；
setDictionaryPath（“resource:/edu/cmu/sphinx/models/en-us/cmudict en-us.dict”）；
setLanguageModelPath（“resource:/edu/cmu/sphinx/models/en-us/en-us.lm.bin”）；
setUseGrammar（false）；
LiveSpeechRecognizer recognizer=新的LiveSpeechRecognizer（配置）；
识别器。startRecognition（真）；
System.out.println（“开始讲话。\n”）；
演讲结果；
长定时器=0；
同时（定时器<2*60*1000）
{
结果=识别器。getResult（）；
如果（结果！=null）
{
系统输出打印（“你说：”）；
for（WordResult-word:result.getWords（））
{
如果（！word.isFiller（））
System.out.print（word.getWord（）+）；
}
System.out.println（）；
}
其他的
{
System.out.println（“我听不清你说了什么。\n”）；
}
}
识别器。停止识别（）；
}
}

谢谢你，seb，但现在的输出就像你说的：那就是我只需要这就是如何避免这种情况，好吧，这是一个填充标记。你需要避免打印这些。我会在那里编辑。我添加了一个复选框，所以我们只在这个词不是填充词的情况下打印它。是的，现在它正在工作，但准确率为0%。你能给我一些建议来提高准确率吗？]你应该在这方面发布一个新问题，我不确定斯芬克斯是如何工作的。你可能需要做一些机器学习。

You said: i am a java student

public class Indian
{
    public static void main(String[] args) throws Exception
    {
        Configuration configuration = new Configuration();
        configuration.setAcousticModelPath("resource:/edu/cmu/sphinx/models/en-us/en-us");
        configuration.setDictionaryPath("resource:/edu/cmu/sphinx/models/en-us/cmudict-en-us.dict");
        configuration.setLanguageModelPath("resource:/edu/cmu/sphinx/models/en-us/en-us.lm.bin");
        configuration.setUseGrammar(false);

        LiveSpeechRecognizer recognizer = new LiveSpeechRecognizer(configuration);

        recognizer.startRecognition(true);
        System.out.println("Start speaking. \n");
        SpeechResult result;
        long timer = 0;

        while (timer < 2 * 60 * 1000)
        {
            result = recognizer.getResult();
            if (result != null)
            {
                System.out.print("You said: ");
                for (WordResult word : result.getWords())
                {
                    if (!word.isFiller())
                        System.out.print(word.getWord() + " ");
                }
                System.out.println();
            }
            else
            {
                System.out.println("I can't hear what you said.\n");
            }
        }
        recognizer.stopRecognition();
    }
}