When the response lenght is long, it will be chunked to multiple part…

…s, and need multiple times of read to receive the whole stream

PronunciationAssessment/Java/jre/src/com/microsoft/cognitiveservices/pronunciationassessment/Sample.java

@@ -93,7 +93,14 @@ public static void main(String[] args) throws Exception {
 		// receive response
 		byte[] responseBuffer = new byte[connection.getContentLength()];
 		InputStream inputStream = connection.getInputStream();
-		inputStream.read(responseBuffer);
+		int offset = 0;
+		int readBytes = inputStream.read(responseBuffer);
+		while (readBytes != -1)
+		{
+			offset += readBytes;
+			readBytes = inputStream.read(responseBuffer, offset, responseBuffer.length - offset);
+		}
+
 		String result = new String(responseBuffer, "utf-8"); // the result is a JSON, you can parse it with a JSON library 
 
 		System.out.println("Pronunciation assessment result:\n");