I just tried it, and it worked for me.
However understand a lot of character to audios may be inaccurate, because of characters with multiple pronunciations.
If the system doesnt have a good dictionary or able to detect words within that string that contain different pronunciations, the audio will come out wrong.
One other method is to break down sentences, and connect the right pronunciation through its jyutping. I have a whole bunch of audios with jyutping that maybe of use within that context, but they need to be programmatically connected.
However understand a lot of character to audios may be inaccurate, because of characters with multiple pronunciations.
If the system doesnt have a good dictionary or able to detect words within that string that contain different pronunciations, the audio will come out wrong.
One other method is to break down sentences, and connect the right pronunciation through its jyutping. I have a whole bunch of audios with jyutping that maybe of use within that context, but they need to be programmatically connected.