Model save
Browse files- README.md +12 -8
- model.safetensors +1 -1
README.md
CHANGED
|
@@ -1,7 +1,5 @@
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
| 3 |
-
language:
|
| 4 |
-
- en
|
| 5 |
license: apache-2.0
|
| 6 |
base_model: openai/whisper-small
|
| 7 |
tags:
|
|
@@ -9,19 +7,19 @@ tags:
|
|
| 9 |
metrics:
|
| 10 |
- wer
|
| 11 |
model-index:
|
| 12 |
-
- name: Whisper-squeezeformer-
|
| 13 |
results: []
|
| 14 |
---
|
| 15 |
|
| 16 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
|
| 19 |
-
# Whisper-squeezeformer-
|
| 20 |
|
| 21 |
-
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
-
- Loss: 0.
|
| 24 |
-
- Wer:
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
|
@@ -47,7 +45,7 @@ The following hyperparameters were used during training:
|
|
| 47 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
- lr_scheduler_type: linear
|
| 49 |
- lr_scheduler_warmup_steps: 2500
|
| 50 |
-
- training_steps:
|
| 51 |
- mixed_precision_training: Native AMP
|
| 52 |
|
| 53 |
### Training results
|
|
@@ -66,6 +64,12 @@ The following hyperparameters were used during training:
|
|
| 66 |
| 0.1467 | 10.0 | 25000 | 0.1494 | 6.9671 |
|
| 67 |
| 0.1113 | 11.0 | 27500 | 0.1493 | 6.7597 |
|
| 68 |
| 0.0914 | 12.0 | 30000 | 0.1511 | 6.8035 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 69 |
|
| 70 |
|
| 71 |
### Framework versions
|
|
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
|
|
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: openai/whisper-small
|
| 5 |
tags:
|
|
|
|
| 7 |
metrics:
|
| 8 |
- wer
|
| 9 |
model-index:
|
| 10 |
+
- name: Whisper-squeezeformer-v3
|
| 11 |
results: []
|
| 12 |
---
|
| 13 |
|
| 14 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
|
| 17 |
+
# Whisper-squeezeformer-v3
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the None dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.1297
|
| 22 |
+
- Wer: 5.6204
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 45 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
- lr_scheduler_type: linear
|
| 47 |
- lr_scheduler_warmup_steps: 2500
|
| 48 |
+
- training_steps: 45000
|
| 49 |
- mixed_precision_training: Native AMP
|
| 50 |
|
| 51 |
### Training results
|
|
|
|
| 64 |
| 0.1467 | 10.0 | 25000 | 0.1494 | 6.9671 |
|
| 65 |
| 0.1113 | 11.0 | 27500 | 0.1493 | 6.7597 |
|
| 66 |
| 0.0914 | 12.0 | 30000 | 0.1511 | 6.8035 |
|
| 67 |
+
| 0.1946 | 13.0 | 32500 | 0.1391 | 6.4212 |
|
| 68 |
+
| 0.1425 | 14.0 | 35000 | 0.1369 | 5.8753 |
|
| 69 |
+
| 0.1145 | 15.0 | 37500 | 0.1368 | 5.7536 |
|
| 70 |
+
| 0.1776 | 16.0 | 40000 | 0.1302 | 5.5995 |
|
| 71 |
+
| 0.1416 | 17.0 | 42500 | 0.1298 | 5.6204 |
|
| 72 |
+
| 0.1239 | 18.0 | 45000 | 0.1297 | 5.6204 |
|
| 73 |
|
| 74 |
|
| 75 |
### Framework versions
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1128325056
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dab8c0f9575fbbcd5a8bcd8a9c5d9cb115e91097d852c52848262942623e6c72
|
| 3 |
size 1128325056
|