Update README.md

README.md

@@ -5,10 +5,11 @@ language:
 - shi
 - rif
 - tzm
+- shy
 license: cc-by-4.0
 library_name: nemo
 datasets:
-- mozilla-foundation/common_voice_17_0
+- mozilla-foundation/common_voice_18_0
 thumbnail: null
 tags:
 - automatic-speech-recognition
@@ -26,8 +27,8 @@ model-index:
       type: Automatic Speech Recognition
       name: automatic-speech-recognition
     dataset:
-      name: Mozilla Common Voice 17.0
-      type: mozilla-foundation/common_voice_17_0
+      name: Mozilla Common Voice 18.0
+      type: mozilla-foundation/common_voice_18_0
       config: zgh
       split: test
       args:
@@ -35,13 +36,13 @@ model-index:
     metrics:
     - name: Test WER
       type: wer
-      value: 71.88
+      value: 64.17
   - task:
       type: Automatic Speech Recognition
       name: automatic-speech-recognition
     dataset:
-      name: Mozilla Common Voice 17.0
-      type: mozilla-foundation/common_voice_17_0
+      name: Mozilla Common Voice 18.0
+      type: mozilla-foundation/common_voice_18_0
       config: zgh
       split: test
       args:
@@ -49,13 +50,13 @@ model-index:
     metrics:
     - name: Test CER
       type: cer
-      value: 25.02
+      value: 21.54
   - task:
       type: Automatic Speech Recognition
       name: automatic-speech-recognition
     dataset:
-      name: Mozilla Common Voice 17.0
-      type: mozilla-foundation/common_voice_17_0
+      name: Mozilla Common Voice 18.0
+      type: mozilla-foundation/common_voice_18_0
       config: kab
       split: test
       args:
@@ -63,13 +64,13 @@ model-index:
     metrics:
     - name: Test WER
       type: wer
-      value: 38.21
+      value: 34.87
   - task:
       type: Automatic Speech Recognition
       name: automatic-speech-recognition
     dataset:
-      name: Mozilla Common Voice 17.0
-      type: mozilla-foundation/common_voice_17_0
+      name: Mozilla Common Voice 18.0
+      type: mozilla-foundation/common_voice_18_0
       config: kab
       split: test
       args:
@@ -77,7 +78,7 @@ model-index:
     metrics:
     - name: Test CER
       type: cer
-      value: 14.41
+      value: 13.11
 metrics:
 - wer
 - cer
@@ -135,11 +136,11 @@ This model provides transcribed speech as a string for a given audio sample.
 
 ## Training
 
-The model was trained for 48 epochs on a NVIDIA GeForce RTX 4050 Laptop GPU.
+The model was fine-tuned from an older checkpoint on a NVIDIA GeForce RTX 4050 Laptop GPU.
 
 ### Datasets
 
-Common Voice 17 *kab* and *zgh* splits plus bible readings in Tachelhit and Tarifit.
+Common Voice 18 *kab* and *zgh* splits, Tatoeba (kab, ber, shy), and bible readings in Tachelhit and Tarifit.
 
 ## Performance