add new model
Browse files- README.md +12 -0
- pytorch_model_v2.bin +3 -0
README.md
CHANGED
@@ -83,5 +83,17 @@ All the Common Voice `Test` dataset (15763 files) were used for testing.
|
|
83 |
Results:
|
84 |
|
85 |
WER=20.89%
|
|
|
86 |
SER=77.56%
|
87 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
83 |
Results:
|
84 |
|
85 |
WER=20.89%
|
86 |
+
|
87 |
SER=77.56%
|
88 |
|
89 |
+
|
90 |
+
## New Model (v2)
|
91 |
+
|
92 |
+
~10% of the Common Voice `train`, `validation` datasets (30K files) were used for training.
|
93 |
+
|
94 |
+
Results:
|
95 |
+
|
96 |
+
WER=18.81%
|
97 |
+
|
98 |
+
SER=73.82%
|
99 |
+
|
pytorch_model_v2.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:58735873545095dd226e8e8064f2b5bcd09bf29f9de74175a7b0cc7b7d777494
|
3 |
+
size 1262097815
|