Adding   a single voice prediction

Read the source code, found that can not add a single voice prediction. From the issue,I find you are adding this module. May I know the result of your adding? I'm trying my best to implementation it. Can you give me a hint? If someone implementation them, can you share the source code。