dima1234321/wav2vec2-xls-r-300m-phoneme_he1

README.md

@@ -1,6 +1,6 @@
 ---
 license: apache-2.0
-base_model: facebook/wav2vec2-xls-r-300m
+base_model: vitouphy/wav2vec2-xls-r-300m-timit-phoneme
 tags:
 - generated_from_trainer
 metrics:
@@ -15,11 +15,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 # shared_audio
 
-This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the None dataset.
+This model is a fine-tuned version of [vitouphy/wav2vec2-xls-r-300m-timit-phoneme](https://huggingface.co/vitouphy/wav2vec2-xls-r-300m-timit-phoneme) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 19.5815
-- Wer: 1.0
-- Cer: 1.0
+- Loss: 66.1707
+- Wer: 1.2143
+- Cer: 9.2148
 
 ## Model description
 
@@ -46,23 +46,14 @@ The following hyperparameters were used during training:
 - total_train_batch_size: 32
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 20
-- training_steps: 100
+- lr_scheduler_warmup_steps: 1
+- training_steps: 2
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Wer | Cer    |
-|:-------------:|:-----:|:----:|:---------------:|:---:|:------:|
-| 62.6565       | 0.16  | 10   | 63.7309         | 1.0 | 2.9738 |
-| 46.9251       | 0.32  | 20   | 54.7811         | 1.0 | 1.0    |
-| 43.8044       | 0.48  | 30   | 32.9802         | 1.0 | 1.0    |
-| 30.6358       | 0.64  | 40   | 26.5389         | 1.0 | 1.0    |
-| 23.5351       | 0.8   | 50   | 23.3499         | 1.0 | 1.0    |
-| 29.7649       | 0.96  | 60   | 21.5416         | 1.0 | 1.0    |
-| 21.3599       | 1.12  | 70   | 20.5404         | 1.0 | 1.0    |
-| 19.8412       | 1.28  | 80   | 19.8837         | 1.0 | 1.0    |
-| 23.5039       | 1.44  | 90   | 19.5565         | 1.0 | 1.0    |
-| 20.1719       | 1.6   | 100  | 19.4386         | 1.0 | 1.0    |
+| Training Loss | Epoch | Step | Validation Loss | Wer    | Cer    |
+|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
+| 58.4649       | 0.03  | 2    | 66.1707         | 1.2143 | 9.2148 |
 
 
 ### Framework versions

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "facebook/wav2vec2-xls-r-300m",
+  "_name_or_path": "vitouphy/wav2vec2-xls-r-300m-timit-phoneme",
   "activation_dropout": 0.0,
   "adapter_attn_dim": null,
   "adapter_kernel_size": 3,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:83096bb2075586342fa0703c2cf4e1d8c4124779a58c55deb8bc2e65bd53b13b
+oid sha256:0053715938cba964d4ed015b37fd6dedead4c0fa24cec27bad839909e505c594
 size 1262233880

runs/Mar24_18-07-46_c9967808b69b/events.out.tfevents.1711303747.c9967808b69b.1519.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:687845e2e5a8ffc3216523a5f2cac0d368fc5008ab6dec2c903f85c8c3223939
+size 7194

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:92e96185dc6b382024930bfbbd505f2b7aed9db60a71e6eecf8e372096d5d0f2
+oid sha256:63d7a7de20897f48f2c5d7c4cffb613e54f95d6e572b05d78ab8f9c50cf71e1e
 size 4920