Update README.md
Browse files
README.md
CHANGED
@@ -1,5 +1,11 @@
|
|
1 |
---
|
2 |
-
language: sv
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
tags:
|
4 |
- audio
|
5 |
- automatic-speech-recognition
|
@@ -15,11 +21,14 @@ model-index:
|
|
15 |
dataset:
|
16 |
name: Common Voice sv-SE
|
17 |
type: common_voice
|
18 |
-
args: sv
|
19 |
metrics:
|
20 |
- name: Test WER
|
21 |
type: wer
|
22 |
-
value:
|
|
|
|
|
|
|
23 |
---
|
24 |
|
25 |
# Wav2Vec2-Large-XLSR-53-Swedish
|
@@ -84,7 +93,7 @@ processor = Wav2Vec2Processor.from_pretrained("KBLab/wav2vec2-large-xlsr-53-swed
|
|
84 |
model = Wav2Vec2ForCTC.from_pretrained("KBLab/wav2vec2-large-xlsr-53-swedish")
|
85 |
model.to("cuda")
|
86 |
|
87 |
-
chars_to_ignore_regex = '[
|
88 |
resampler = torchaudio.transforms.Resample(48_000, 16_000)
|
89 |
|
90 |
# Preprocessing the datasets.
|
|
|
1 |
---
|
2 |
+
language: sv-SE
|
3 |
+
datasets:
|
4 |
+
- common_voice
|
5 |
+
- NST Swedish ASR Database
|
6 |
+
metrics:
|
7 |
+
- wer
|
8 |
+
- cer
|
9 |
tags:
|
10 |
- audio
|
11 |
- automatic-speech-recognition
|
|
|
21 |
dataset:
|
22 |
name: Common Voice sv-SE
|
23 |
type: common_voice
|
24 |
+
args: sv-SE
|
25 |
metrics:
|
26 |
- name: Test WER
|
27 |
type: wer
|
28 |
+
value: 18.252392
|
29 |
+
- name: Test CER
|
30 |
+
type: cer
|
31 |
+
value: 5.965446
|
32 |
---
|
33 |
|
34 |
# Wav2Vec2-Large-XLSR-53-Swedish
|
|
|
93 |
model = Wav2Vec2ForCTC.from_pretrained("KBLab/wav2vec2-large-xlsr-53-swedish")
|
94 |
model.to("cuda")
|
95 |
|
96 |
+
chars_to_ignore_regex = '[,?.!\\-;:"“]'
|
97 |
resampler = torchaudio.transforms.Resample(48_000, 16_000)
|
98 |
|
99 |
# Preprocessing the datasets.
|