File size: 492 Bytes
aa46b06
 
 
 
 
 
 
 
 
a66fdc5
 
 
 
 
 
2a44917
55607a8
2a44917
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
---
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" course at Tel-Aviv University.

Implementation details and references can be found in the included 'paper' PDF. \
Code can be found on [our GitHub Repo](https://github.com/D4niel0s/HebTTS)