search

alphacep / vosk-api

Offline speech recognition API for Android, iOS, Raspberry Pi and servers with Python, Java, C# and Node
Apache License 2.0
7.35k stars 1.04k forks source link

How to deal with homonyms in grammer #1565

Closed MooWeii closed 1 month ago

MooWeii commented 2 months ago

I defined grammer, in which there are two homophonic Chinese characters "摆" and "百". Now there is a sentence with the correct format of "摆料等待时间修改为一百", but always return "百料等待时间修改为一百". It doesn't work if I add "摆料" to grammer, so what should I do

nshmyrev commented 2 months ago

Hi

You can improve language model to give more priority for correct word. See https://alphacephei.com/vosk/lm

MooWeii commented 2 months ago

Hi

You can improve language model to give more priority for correct word. See https://alphacephei.com/vosk/lm

Thank you for your reply and contribution. In addition, where can I download the Chinese upgrade package. I see that the upgrade needs to download the upgrade package, but there is no Chinese link.

nshmyrev commented 2 months ago

You can request it over email

nshmyrev commented 1 month ago

Same as #866