Model save

README.md

@@ -1,7 +1,7 @@
 ---
 library_name: transformers
 license: apache-2.0
-base_model: openai/whisper-base
+base_model: openai/whisper-medium
 tags:
 - generated_from_trainer
 metrics:
@@ -16,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # whisper-finetuned-maghreb-dialect-detector
 
-This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the None dataset.
+This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.6024
-- Wer: 90.2221
+- Loss: 1.4892
+- Wer: 67.2099
 
 ## Model description
 
@@ -54,10 +54,10 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch   | Step | Validation Loss | Wer     |
 |:-------------:|:-------:|:----:|:---------------:|:-------:|
-| 1.2071        | 3.0770  | 1000 | 1.4660          | 91.4040 |
-| 0.4595        | 6.1541  | 2000 | 1.6024          | 90.2221 |
-| 0.2045        | 9.2311  | 3000 | 1.7459          | 92.2905 |
-| 0.1057        | 12.3082 | 4000 | 1.8226          | 90.4549 |
+| 0.4326        | 3.0770  | 1000 | 1.1722          | 74.2031 |
+| 0.0454        | 6.1541  | 2000 | 1.3672          | 68.7769 |
+| 0.0097        | 9.2311  | 3000 | 1.4892          | 67.2099 |
+| 0.0008        | 12.3082 | 4000 | 1.5572          | 67.4875 |
 
 
 ### Framework versions

generation_config.json

@@ -1,36 +1,28 @@
 {
   "alignment_heads": [
     [
-      3,
-      1
-    ],
-    [
-      4,
-      2
+      13,
+      15
     ],
     [
-      4,
-      3
+      15,
+      4
     ],
     [
-      4,
-      7
+      15,
+      15
     ],
     [
-      5,
+      16,
       1
     ],
     [
-      5,
-      2
+      20,
+      0
     ],
     [
-      5,
+      23,
       4
-    ],
-    [
-      5,
-      6
     ]
   ],
   "begin_suppress_tokens": [

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4a9459882140fe958d3ab2eead3a810e1f8620809227e2aa2244a17bb35d01e2
+oid sha256:f2ebdb78165d27b4b6c1bf4b22a6d3d91abc50b0bc1d066402cbcd12099e6113
 size 3055556772

runs/Mar29_03-02-39_tech-Z790-AORUS-ELITE-AX/events.out.tfevents.1743210160.tech-Z790-AORUS-ELITE-AX.94244.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f5401a6744b943b064f5565a40031a40405950c74498cb96f7c271df73ee6b52
-size 9373
+oid sha256:218712c83458e135f13ead0f6bcad94f03fecdd0c49f01661ee617a697d96841
+size 17011

runs/Mar29_03-02-39_tech-Z790-AORUS-ELITE-AX/events.out.tfevents.1743220848.tech-Z790-AORUS-ELITE-AX.94244.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a88505fdeb835d60264bf846d19092d7708002e8423ac5dab0e6d4be0290a3d
+size 406