HebTTS

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+additional_info.pdf filter=lfs diff=lfs merge=lfs -text
+dataset/2407.07566v1[[:space:]]HEBDB.[[:space:]]A[[:space:]]Weakly[[:space:]]Supervised[[:space:]]Dataset[[:space:]]for[[:space:]]Hebrew[[:space:]]Speech[[:space:]]Processing.pdf filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,18 @@
+---
+datasets:
+- SLPRL-HUJI/HebDB
+language:
+- he
+metrics:
+- wer
+- cer
+pipeline_tag: text-to-speech
+---
+
+
+# Details
+
+This model is an implementation of the vall-e architecture, with the AlephBert text tokenizer.
+This model was trained as a final project in the "DSP & audio processing using Deep Learning" class at Tel-Aviv University, Israel.
+
+Implementation details and references can be found in the included 'paper' PDF.

additional_info.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89b412730083c73688f540f0358bc1808ef038baf33f27df7e729e0ec7e8f9ec
+size 509426

checkpoint-150000.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1fecc21e683dd103a7bfed57f986c14c84213f93aec815af73947e3dfec63e81
+size 2551217450

dataset/2407.07566v1 HEBDB. A Weakly Supervised Dataset for Hebrew Speech Processing.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97febbefdb88fd63147d9eec9bc60138480a78a44eb0d40061bdce2e26a6cfa8
+size 278562

source.txt

@@ -0,0 +1,2 @@
+https://huggingface.co/D4niel0s/HebTTS_implementation
+https://huggingface.co/datasets/SLPRL-HUJI/HebDB