java - 如何使用 java 和 CMU Sphinx 将音频文件分解为多个片段，然后将分解的音频文件转录为文本

Question

我已经编写了一个代码，可以将音频文件转录为文本，但我的问题是我想将音频文件分成几部分，然后我想一个接一个地转录该中断的音频文件，请帮助我

        StreamSpeechRecognizer recognizer;
        try
        {
            recognizer = new StreamSpeechRecognizer( configuration);
            java.io.InputStream stream = AppRunner.class.getResourceAsStream(splitFile(new File("/com/dsquare/Arabtec_Construction_INDIA_Private_Limited_convert.wav")));

            System.out.println(stream);
            stream.skip(44);

            // Simple recognition with generic model
            recognizer.startRecognition(stream);
            SpeechResult result;
            while ((result = recognizer.getResult()) != null) 
            {

             System.out.format("Hypothesis: %s\n", result.getHypothesis());

             System.out.println("List of recognized words and their times:");
             for (WordResult r : result.getWords()) 
             {
             System.out.println(r);
             }

            // System.out.println("Best 3 hypothesis:");
             for (String s : result.getNbest(3))
             {
         System.out.println(s);
          }
            recognizer.stopRecognition();

        }
        }
        catch (IOException e)
        {
            // TODO Auto-generated catch block
            e.printStackTrace();
        }







    public static String splitFile(File f) throws IOException {
        int partCounter = 1;//I like to name parts from 001, 002, 003, ...
                            //you can change it to 0 if you want 000, 001, ...

        int sizeOfFiles = 1024 * 1024;// 1MB
        byte[] buffer = new byte[sizeOfFiles];

        try (BufferedInputStream bis = new BufferedInputStream(
                new FileInputStream(f))) {//try-with-resources to ensure closing stream
            String name = f.getName();

            int tmp = 0;
            while ((tmp = bis.read(buffer)) > 0) {
                //write each chunk of data into separate file with different number in name
                File newFile = new File(f.getParent(), name + "."
                        + String.format("%03d", partCounter++));
                try (FileOutputStream out = new FileOutputStream(newFile)) {
                    out.write(buffer, 0, tmp);//tmp is chunk size
                }
            }
        }
        return null;
    }

}

Answer 1

要以智能方式破坏音频文件，您可以考虑分类工具，例如 Lium 小组开发的这个工具。

http://www-lium.univ-lemans.fr/diarization/doku.php/welcome

此工具将为您提供一个包含过渡时间的 *.seg 文件。然后，使用ffmpeg或类似工具来剪切文件。