mandarin dict is empty without any error

[Nuller-CV]

When I use the my pretrained G2P model and example Mandarin corpus(from Example 2: Generate Mandarin dictionary), I got the empty mandrin_dict.txt without any error:

Generating pronunciations from G2P model Generating transcriptions for the 102 word types found in the corpus... WARNING! The following graphemes were not found in the specified G2P model: 1 2 3 4 5 a b c d e f g h i j k l m n o p q r s t u v w x y z Generating pronunciations... 0%| | 0/102 [00:00<?, ?it/s] Processed 102 in 1.0486838817596436 seconds

[mmcauliffe]

I just tested the steps and it's generating correctly for me. From the error message, it might be the case that you're using the character based G2P model? Could you try rerunning mfa download g2p mandarin_pinyin_g2p. Also maybe trying upgrading to the latest MFA version (pip install montreal-forced-aligner -U, and conda upgrade -c conda-forge openfst pynini ngram baumwelch), just to eliminate any potential bugs that I have fixed recently.