Closed Wu-Urbanek closed 4 years ago
Yes. ⁶ and ⁸ are no real tones, but RELATIVE tones.
We have that, unfortunately, in some datasets.
The workaround is, to write all tones with the /
construct:
Grapheme | IPA |
---|---|
¹ | ¹/¹ |
² | ²/² |
... | ... |
⁶ | ⁶/⁵¹ |
⁷ | ⁷/⁵² |
⁸ | ⁸/⁵³ |
as you can see: we distinguish the tones in the end, but we do as if they are "pronounced", although there is no real pronunciation given in the data...
I see. I modified the tones ( 6 and 8 ). I will refine it further if we can get the PDF file "Vocabularies of Guangxi ethnic languages[广西民族语言方音词汇]".
There are two BIPA errors now. The 6 and 8 should be tones, I replaced "6" to "⁶" and "8" to "⁸". Any suggestion?