Timothy Mak

Error with v2.0.0b8/v2.0.0b9

11

Hi I'm encountering the following error. The setup is identical to #359, except that I'm setting the `subset` option to 0. ``` Number of jobs starting: 1 DEBUG - Set...

Separate the MFA source codes from the dependencies in the conda package

6

Would it be possible to separate the source codes (i.e. what's in this repository) from the dependencies (kaldi, etc.) in the conda MFA package? In this way, it will be...

I think if you want your data to be non-commercial, the license should be CC-BY-NC (https://creativecommons.org/licenses/by-nc/4.0/) rather than CC-BY (https://creativecommons.org/licenses/by/4.0/).

Strange attention_weights matrix

17

I'm training the flowtron model from scratch on the LJSpeech dataset. It seems to run ok. However, after nearly three days, the attention matrix still has the following form and...

Formula for converting PitchTier to Point Process

2

Hi, I am trying to understand the maths behind Praat. Would you be able to explain where the algorithm [here](https://github.com/praat/praat/blob/382c64e43c64bf73b93fcec32ebfd788b5970a8d/fon/PitchTier_to_PointProcess.cpp#L39) comes from? Many thanks! Tim

I've upgraded my MFA to version v2.2.12. However, for my downstream task (TTS), my old MFA version (v2.0.0a23) seems to perform better. What are the major changes since v2.0.0a23 that...

How can I specify the name of the subdirectory used within the <temporary_directory>?

2

This is more of a question. Is there an option to specify the sub-directory used within ? Currently, it seems to default to /. However, I'd like to be able...

Wrong parsing of AISHELL3 text

3

I see that on line https://github.com/DigitalPhonetics/IMS-Toucan/blob/v2.5/Utility/path_to_transcript_dicts.py#L478, you are treating the "%" characters in the AISHELL3 transcript as if they are commas. However, they are actually "word delimiters" and do not...

Duplicate entries in jyutping dictionary

2

Hi, I found that some characters, e.g. 凹 has duplicate entry in the dictionary: ``` 凹 凹 au3/nap1 凹 凹 waa1 ``` leading to incomplete results when we try to...

Wav2Vec2CTCTokenizer adds random unknown tokens to encoded input

2

### System Info - `transformers` version: 4.29.2 - Platform: Linux-5.15.0-92-generic-x86_64-with-glibc2.35 - Python version: 3.11.8 - Huggingface_hub version: 0.22.2 - Safetensors version: 0.4.2 - PyTorch version (GPU?): 2.2.2 (True) - Tensorflow...