wav2vec2-large-mms-1b-urmi-christian

README.md

@@ -32,9 +32,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the nena_speech_1_0_test dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9610
+- Loss: 1.0957
 - Wer: 1.0
-- Cer: 0.3016
+- Cer: 0.3080
 
 ## Model description
 
@@ -66,7 +66,8 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Wer | Cer    |
 |:-------------:|:-----:|:----:|:---------------:|:---:|:------:|
-| 4.7963        | 3.12  | 100  | 0.9610          | 1.0 | 0.3016 |
+| 8.123         | 1.56  | 50   | 3.2235          | 1.0 | 0.9678 |
+| 1.9948        | 3.12  | 100  | 1.0957          | 1.0 | 0.3080 |
 
 
 ### Framework versions

adapter.aii.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d38e1572fcc670d481aa52489983274bf1a86d3557b7ff8591d3eabaf7b3db2a
+oid sha256:507d3de2f115903f6ce975b292443231ff352290e6f1c8c4770f0bd25e452f32
 size 8870276

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:87928e6cce5fbe29119d4652d76dce6e4cbadb268ea35f18ce591d5b8153ae1f
+oid sha256:bdaa2467a07988e8d233fcaa294177562f031733289c674a0a9c2a59a15d1ad3
 size 4536