-
Notifications
You must be signed in to change notification settings - Fork 52
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
symbol not found in flat namespace '__ZN5MeCab11createModelEPKc' #87
Comments
This is definitely not caused by a bug in tts. However if you aren't using mecab-python3 (like, you're not working with Japanese), they shouldn't automatically install it, and it's unfortunate that they do so. Thank you for bringing this to my attention. I am not sure exactly what mismatch causes this error. You're building
If you installed Also, to clarify something for me, does brew build MeCab from source, or does it install a precompiled binary? Depending on how that works it could cause different kinds of issues. Can you load MeCab on its own, like this? (You'll get an error without a dictionary, but that'll show there's no symbol error at least.)
|
Thanks also for posting on the TTS repo to suggest removing the mecab dependency!
Almost correct. On the Apple Silicon (arm64) architecture, you need to
I'm not sure either why this happens. Yes I did install
That is a good point I will check, thanks. As for pip, I made sure it compiles from source with:
Interesting!
So I will try to build homebrew from source and let you know soon. |
@polm you're a genius!
and then rerun the pip install from source:
Result: Thank you indeed. |
Thanks for confirming it worked, glad you got it running. |
Last addon: |
unfortunately it still does not work for me - I have installed mecab-python3 as per the suggestion, but if I use |
I have filed this issue on the tts github as bug here, but I think it can only be solved here as the error is thrown from the mecab library.
Please note that the error is thrown regardless which model or vocoder is used.
Example:
tts --text "Text for TTS" --model_name "tts_models/en/ek1/tacotron2" --vocoder_name "vocoder_models/en/vctk/hifigan_v2" --out_path output/speech.wav
throws this
Question
Could this error be related to the correct version of
boost-python3
used for compiling, as is explained thoroughly here?The text was updated successfully, but these errors were encountered: