Commit
•
bcfe9f7
1
Parent(s):
16b0c23
End of training
Browse files
README.md
CHANGED
|
@@ -7,21 +7,9 @@ tags:
|
|
| 7 |
- generated_from_trainer
|
| 8 |
datasets:
|
| 9 |
- jayasuryajsk/google-fleurs-te-romanized
|
| 10 |
-
metrics:
|
| 11 |
-
- wer
|
| 12 |
model-index:
|
| 13 |
- name: Wishper-Large-V3-spoken_telugu_romanized
|
| 14 |
-
results:
|
| 15 |
-
- task:
|
| 16 |
-
name: Automatic Speech Recognition
|
| 17 |
-
type: automatic-speech-recognition
|
| 18 |
-
dataset:
|
| 19 |
-
name: Telugu Romanized 1.0
|
| 20 |
-
type: jayasuryajsk/google-fleurs-te-romanized
|
| 21 |
-
metrics:
|
| 22 |
-
- name: Wer
|
| 23 |
-
type: wer
|
| 24 |
-
value: 67.44785563627842
|
| 25 |
---
|
| 26 |
|
| 27 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -31,8 +19,13 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 31 |
|
| 32 |
This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the Telugu Romanized 1.0 dataset.
|
| 33 |
It achieves the following results on the evaluation set:
|
| 34 |
-
-
|
| 35 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 36 |
|
| 37 |
## Model description
|
| 38 |
|
|
@@ -61,14 +54,6 @@ The following hyperparameters were used during training:
|
|
| 61 |
- training_steps: 2000
|
| 62 |
- mixed_precision_training: Native AMP
|
| 63 |
|
| 64 |
-
### Training results
|
| 65 |
-
|
| 66 |
-
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 67 |
-
|:-------------:|:-------:|:----:|:---------------:|:-------:|
|
| 68 |
-
| 0.0079 | 8.6207 | 1000 | 1.4603 | 64.9051 |
|
| 69 |
-
| 0.0007 | 17.2414 | 2000 | 1.5824 | 67.4479 |
|
| 70 |
-
|
| 71 |
-
|
| 72 |
### Framework versions
|
| 73 |
|
| 74 |
- Transformers 4.40.1
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
datasets:
|
| 9 |
- jayasuryajsk/google-fleurs-te-romanized
|
|
|
|
|
|
|
| 10 |
model-index:
|
| 11 |
- name: Wishper-Large-V3-spoken_telugu_romanized
|
| 12 |
+
results: []
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
---
|
| 14 |
|
| 15 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 19 |
|
| 20 |
This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the Telugu Romanized 1.0 dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
+
- eval_loss: 1.5009
|
| 23 |
+
- eval_wer: 68.1275
|
| 24 |
+
- eval_runtime: 591.6137
|
| 25 |
+
- eval_samples_per_second: 0.798
|
| 26 |
+
- eval_steps_per_second: 0.1
|
| 27 |
+
- epoch: 8.6207
|
| 28 |
+
- step: 1000
|
| 29 |
|
| 30 |
## Model description
|
| 31 |
|
|
|
|
| 54 |
- training_steps: 2000
|
| 55 |
- mixed_precision_training: Native AMP
|
| 56 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 57 |
### Framework versions
|
| 58 |
|
| 59 |
- Transformers 4.40.1
|
model-00001-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4993448880
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:85faa737292ae66b15ee0d82299ff9ac691a6f4ee3d631f3cc180b010106de0c
|
| 3 |
size 4993448880
|
model-00002-of-00002.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1180663192
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b5b84dacdfc8c6730a8f8b4285f74a77f5bb917d4c9c24dcbc281c2032271337
|
| 3 |
size 1180663192
|
runs/May07_02-44-27_36aefd7941c6/events.out.tfevents.1715049870.36aefd7941c6.6372.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:04d47026169fa0c492b647ececca5c476c3b51c5cb5bbb8b99d623366a6115f6
|
| 3 |
+
size 268879
|