Add support for the Hubert CTC model (#5)

* 🚧 add HubertForCTC for Hubert models * 🚸 add short names for cmd args * 🙈 add VS code * ♻️ move loading model to function * 📝 format & add more quickstart details * ✨ update base model to Hubert * 📝 clean up docstrings * ♻️ clean up docstrings and comments 📝 * 🎨 format to black * 🙈 add common video formats * ♻️ clean up imports * ⚰️ remove obsolete fn * 📝 update TODO items * 🚚 clarify example model in dirname * 📝 clarify FFMPEG req * 🚚 rename to clearer name * ✨ add hubert and streamline * 📝 add instructions for multi file Colab * 📝 update summarization recs * 🚚 rename for clarity rename the single file demo notebook * ✨ add huBERT model + streamline * 🚸 standardize video, clean up * ;memo: update speech and demo notebook * :memo add and cite video itself for 😎 points * 📝 clean up examples * 📝 progress update * ✏️ formatting * 📝 add huBERT cite Co-authored-by: peter szemraj <[email protected]> Co-authored-by: Peter Szemraj <[email protected]>
pszemraj · Jan 28, 2022 · 309c420 · 309c420
1 parent 8f05665
commit 309c420
Show file tree

Hide file tree

Showing 50 changed files with 18,718 additions and 11,194 deletions.
diff --git a/.gitignore b/.gitignore
@@ -133,6 +133,8 @@ pythonenv*
 .dmypy.json
 dmypy.json
 
+# IDE
+.vscode
 
 # pytype static type analyzer
 .pytype/
@@ -155,3 +157,9 @@ scratch
 notebooks
 *audio_chunks*
 *test2*
+
+# media files
+*.mp4
+*.wav
+*.mov
+*.mp3
diff --git a/README.md b/README.md
diff --git a/colab_notebooks/vid2cleantext_multi.ipynb b/colab_notebooks/vid2cleantext_multi.ipynb