Updated README
Browse files
README.md
CHANGED
@@ -8,6 +8,46 @@ tags:
|
|
8 |
- pytorch
|
9 |
- phoneme-recognition
|
10 |
pipeline_tag: automatic-speech-recognition
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
---
|
12 |
|
13 |
Model Information
|
@@ -30,7 +70,13 @@ Note that our baseline models were trained without phonetic feature classifiers
|
|
30 |
Usage
|
31 |
=====
|
32 |
|
33 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
34 |
|
35 |
```python
|
36 |
from allophant.estimator import Estimator
|
@@ -102,3 +148,4 @@ Citation
|
|
102 |
booktitle={{Proc. Interspeech 2023}},
|
103 |
month={8}}
|
104 |
```
|
|
|
|
8 |
- pytorch
|
9 |
- phoneme-recognition
|
10 |
pipeline_tag: automatic-speech-recognition
|
11 |
+
arxiv: arxiv.org/abs/2306.04306
|
12 |
+
metrics:
|
13 |
+
- per
|
14 |
+
- aer
|
15 |
+
library_name: allophant
|
16 |
+
language:
|
17 |
+
- bn
|
18 |
+
- ca
|
19 |
+
- cs
|
20 |
+
- cv
|
21 |
+
- da
|
22 |
+
- de
|
23 |
+
- el
|
24 |
+
- en
|
25 |
+
- es
|
26 |
+
- et
|
27 |
+
- eu
|
28 |
+
- fi
|
29 |
+
- fr
|
30 |
+
- ga
|
31 |
+
- hi
|
32 |
+
- hu
|
33 |
+
- id
|
34 |
+
- it
|
35 |
+
- ka
|
36 |
+
- ky
|
37 |
+
- lt
|
38 |
+
- mt
|
39 |
+
- nl
|
40 |
+
- pl
|
41 |
+
- pt
|
42 |
+
- ro
|
43 |
+
- ru
|
44 |
+
- sk
|
45 |
+
- sl
|
46 |
+
- sv
|
47 |
+
- sw
|
48 |
+
- ta
|
49 |
+
- tr
|
50 |
+
- uk
|
51 |
---
|
52 |
|
53 |
Model Information
|
|
|
70 |
Usage
|
71 |
=====
|
72 |
|
73 |
+
Install the [`allophant`](https://github.com/kgnlp/allophant) package:
|
74 |
+
|
75 |
+
```bash
|
76 |
+
pip install allophant
|
77 |
+
```
|
78 |
+
|
79 |
+
A pre-trained model can be loaded from a huggingface checkpoint or local file:
|
80 |
|
81 |
```python
|
82 |
from allophant.estimator import Estimator
|
|
|
148 |
booktitle={{Proc. Interspeech 2023}},
|
149 |
month={8}}
|
150 |
```
|
151 |
+
[](arxiv.org/abs/2306.04306)
|