The [bird] told you to reject the evidence of your eyes and ears. It was their final, most essential command.
Not a plea for help since I imagine using the word bank allows the text to be accepted. And no, I'm not trying to get it to fail by using 二 instead of ニ. I'm pretty sure the exercise is bugged.
Just a guess, but was there an extra space after the comma? Unlike in English, the full-width comma takes up an entire square worth of space like all other characters and shouldn't have an extra space after it. I don't know if Duolingo even considers spaces when marking answers though so that may not be it.
Although I can't prove it with just the image, my input method editor (fcitx5) seems to only use full-width characters, even for commas, when I use it to input Japanese text. I had completed several dozen other exercises, many containing commas, which seemed to work fine with Duolingo. Even copy-pasting their “Correct solution” did not work and I believe it contained a full-width comma.
If you're familar with Python, can you try this script?
(please replace the string a and b with your input and the answer)
$ python
>>> import unicodedata
>>> a = 'はい、 コンビニです。'
>>> print('\n'.join(f'{hex(ord(c))}, {unicodedata.name(c)}' for c in a))
0x306f, HIRAGANA LETTER HA
0x3044, HIRAGANA LETTER I
0x3001, IDEOGRAPHIC COMMA
0x3000, IDEOGRAPHIC SPACE
0x30b3, KATAKANA LETTER KO
0x30f3, KATAKANA LETTER N
0x30d3, KATAKANA LETTER BI
0x30cb, KATAKANA LETTER NI
0x3067, HIRAGANA LETTER DE
0x3059, HIRAGANA LETTER SU
0x3002, IDEOGRAPHIC FULL STOP
>>> b = 'はい、コンビニです。'
>>> print('\n'.join(f'{hex(ord(c))}, {unicodedata.name(c)}' for c in b))
0x306f, HIRAGANA LETTER HA
0x3044, HIRAGANA LETTER I
0x3001, IDEOGRAPHIC COMMA
0x30b3, KATAKANA LETTER KO
0x30f3, KATAKANA LETTER N
0x30d3, KATAKANA LETTER BI
0x30cb, KATAKANA LETTER NI
0x3067, HIRAGANA LETTER DE
0x3059, HIRAGANA LETTER SU
0x3002, IDEOGRAPHIC FULL STOP