update dependencies and doc accordingly; cleanup wav

Author: slegroux

.gitattributes

@@ -5,3 +5,4 @@
 *.kenlm filter=lfs diff=lfs merge=lfs -text
 *.arpa filter=lfs diff=lfs merge=lfs -text
 *ubyte filter=lfs diff=lfs merge=lfs -text
+*.ipynb merge=nbdev-merge

.gitignore

@@ -133,6 +133,7 @@ recipes/*/logs
 tests/
 .hydra/
 *.png
+*.wav
 *.pt
 *.onnx
 *.ckpt

README.md

@@ -19,9 +19,19 @@ you need python \<3.12
 
 ### Install using Pip
 
+Install package:
 ``` sh
 pip install slg-nimrod
 ```
+Install espeak for LM:
+```bash
+brew install espeak #macos
+```
+Install Spacy english model
+```bash
+python -m spacy download en_core_web_sm
+```
+
 
 ## Usage
 
@@ -33,7 +43,9 @@ git lfs install
 # update changes
 git lfs fetch --all
 # copy the actual data
-git lfs pull
+git lfs checkout
+# or just
+git lfs pull # combing both steps above into one (like usual git pull)
 ```
 
 Check recipes in `recipes/` folder. E.g. for a simple digit recognizer

environment-macos.yml

@@ -5,33 +5,42 @@ channels:
   - conda-forge
   - huggingface
 dependencies:
+  - anyascii
+  - datasets
+  - h5py
+  - hydra-core
+  - inflect
   - jupyterlab
   - matplotlib
-  - tqdm
-  - pandas
-  - datasets
-  - transformers
   - nbdev
-  - spacy
-  - inflect
-  - anyascii
   - omegaconf
-  - hydra-core
-  - wandb
   - onnxruntime
-  - h5py
+  - pandas
+  - portalocker
   - pre-commit
   - pytest
+  - python-dotenv
   - portalocker
+  - spacy
+  - streamlit
+  - optuna
+  - tensorboard
+  - transformers
+  - tqdm
+  - wandb
   - pip:
+    - encodec
+    - hydra_colorlog
+    - hydra-optuna-sweeper
     - jupyterlab-quarto
+    - kenlm
+    - lightning
+    - lhotse
+    - plum-dispatch
+    - https://github.com/kpu/kenlm/archive/master.zip
+    - phonemizer
     - torch
     - torchvision
     - torchtext
+    - torchdata
     - torchaudio
-    - plum-dispatch
-    - https://github.com/kpu/kenlm/archive/master.zip
-    - lhotse
-    - phonemizer
-    - hydra_colorlog
-    - lightning

settings.ini

@@ -37,13 +37,22 @@ language = English
 status = 3
 
 ### Optional ###
-requirements = torch==2.0.1 torchvision==0.15.2 torchtext==0.15.2 torchdata==0.6.1 torchaudio==2.0.2 \
-    anyascii datasets encodec h5py hydra-core hydra-optuna-sweeper inflect jupyterlab==4.0.6 \
+; requirements = torch==2.0.1 torchvision==0.15.2 torchtext==0.15.2 torchdata==0.6.1 torchaudio==2.0.2 \
+;     anyascii datasets encodec h5py hydra-core hydra-optuna-sweeper inflect jupyterlab==4.0.6 \
+;     lightning matplotlib nbdev omegaconf onnxruntime pandas portalocker \
+;     pre-commit python-dotenv pytest spacy==3.6.1 tensorboard tqdm transformers wandb \ 
+;     hydra_colorlog jupyterlab-quarto kenlm lhotse==1.16.0 plum-dispatch \
+;     phonemizer==3.2.1 streamlit optuna 
+;     # https://github.com/kpu/kenlm/archive/master.zip cupy 
+
+requirements = torch torchvision torchtext torchdata torchaudio \
+    anyascii datasets encodec h5py hydra-core hydra-optuna-sweeper inflect jupyterlab \
     lightning matplotlib nbdev omegaconf onnxruntime pandas portalocker \
-    pre-commit python-dotenv pytest spacy==3.6.1 tensorboard tqdm transformers wandb \ 
-    hydra_colorlog jupyterlab-quarto kenlm lhotse==1.16.0 plum-dispatch \
-    phonemizer==3.2.1 streamlit optuna 
+    pre-commit python-dotenv pytest spacy tensorboard tqdm transformers wandb \ 
+    hydra_colorlog jupyterlab-quarto kenlm lhotse plum-dispatch \
+    phonemizer streamlit optuna 
     # https://github.com/kpu/kenlm/archive/master.zip cupy 
 
+
 # dev_requirements = 
 # console_scripts =