End of training
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- README.md +74 -0
- adapter_config.json +32 -0
- adapter_model.safetensors +3 -0
- audio/1.mp3 +0 -0
- audio/10.mp3 +0 -0
- audio/11.mp3 +0 -0
- audio/12.mp3 +0 -0
- audio/13.mp3 +0 -0
- audio/14.mp3 +0 -0
- audio/15.mp3 +0 -0
- audio/16.mp3 +0 -0
- audio/17.mp3 +0 -0
- audio/18.mp3 +0 -0
- audio/19.mp3 +0 -0
- audio/2.mp3 +0 -0
- audio/20.mp3 +0 -0
- audio/21.mp3 +0 -0
- audio/22.mp3 +0 -0
- audio/23.mp3 +0 -0
- audio/24.mp3 +0 -0
- audio/25.mp3 +0 -0
- audio/26.mp3 +0 -0
- audio/27.mp3 +0 -0
- audio/28.mp3 +0 -0
- audio/29.mp3 +0 -0
- audio/3.mp3 +0 -0
- audio/30.mp3 +0 -0
- audio/31.mp3 +0 -0
- audio/32.mp3 +0 -0
- audio/33.mp3 +0 -0
- audio/34.mp3 +0 -0
- audio/35.mp3 +0 -0
- audio/36.mp3 +0 -0
- audio/37.mp3 +0 -0
- audio/38.mp3 +0 -0
- audio/39.mp3 +0 -0
- audio/4.mp3 +0 -0
- audio/40.mp3 +0 -0
- audio/41.mp3 +0 -0
- audio/42.mp3 +0 -0
- audio/43.mp3 +0 -0
- audio/44.mp3 +0 -0
- audio/45.mp3 +0 -0
- audio/46.mp3 +0 -0
- audio/47.mp3 +0 -0
- audio/48.mp3 +0 -0
- audio/49.mp3 +0 -0
- audio/5.mp3 +0 -0
- audio/50.mp3 +0 -0
- audio/6.mp3 +0 -0
README.md
ADDED
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
base_model: openai/whisper-small
|
3 |
+
datasets:
|
4 |
+
- mozilla-foundation/common_voice_11_0
|
5 |
+
language:
|
6 |
+
- hi
|
7 |
+
library_name: peft
|
8 |
+
license: apache-2.0
|
9 |
+
tags:
|
10 |
+
- generated_from_trainer
|
11 |
+
model-index:
|
12 |
+
- name: Whisper Small Diny
|
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 Small Diny
|
20 |
+
|
21 |
+
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 11.0 dataset.
|
22 |
+
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 1.3108
|
24 |
+
|
25 |
+
## Model description
|
26 |
+
|
27 |
+
More information needed
|
28 |
+
|
29 |
+
## Intended uses & limitations
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Training and evaluation data
|
34 |
+
|
35 |
+
More information needed
|
36 |
+
|
37 |
+
## Training procedure
|
38 |
+
|
39 |
+
### Training hyperparameters
|
40 |
+
|
41 |
+
The following hyperparameters were used during training:
|
42 |
+
- learning_rate: 0.0001
|
43 |
+
- train_batch_size: 8
|
44 |
+
- eval_batch_size: 8
|
45 |
+
- seed: 42
|
46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
47 |
+
- lr_scheduler_type: linear
|
48 |
+
- lr_scheduler_warmup_steps: 50
|
49 |
+
- num_epochs: 10
|
50 |
+
- mixed_precision_training: Native AMP
|
51 |
+
|
52 |
+
### Training results
|
53 |
+
|
54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
55 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
56 |
+
| 3.5747 | 1.0 | 4 | 3.8772 |
|
57 |
+
| 3.5154 | 2.0 | 8 | 3.7508 |
|
58 |
+
| 3.3537 | 3.0 | 12 | 3.5159 |
|
59 |
+
| 3.1336 | 4.0 | 16 | 3.2003 |
|
60 |
+
| 2.9481 | 5.0 | 20 | 2.8626 |
|
61 |
+
| 2.5179 | 6.0 | 24 | 2.5561 |
|
62 |
+
| 2.3247 | 7.0 | 28 | 2.2656 |
|
63 |
+
| 2.1071 | 8.0 | 32 | 1.9826 |
|
64 |
+
| 1.7406 | 9.0 | 36 | 1.6799 |
|
65 |
+
| 1.5765 | 10.0 | 40 | 1.3108 |
|
66 |
+
|
67 |
+
|
68 |
+
### Framework versions
|
69 |
+
|
70 |
+
- PEFT 0.12.0
|
71 |
+
- Transformers 4.42.4
|
72 |
+
- Pytorch 2.3.1+cu121
|
73 |
+
- Datasets 2.20.0
|
74 |
+
- Tokenizers 0.19.1
|
adapter_config.json
ADDED
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": {
|
4 |
+
"base_model_class": "WhisperForConditionalGeneration",
|
5 |
+
"parent_library": "transformers.models.whisper.modeling_whisper"
|
6 |
+
},
|
7 |
+
"base_model_name_or_path": "openai/whisper-small",
|
8 |
+
"bias": "none",
|
9 |
+
"fan_in_fan_out": false,
|
10 |
+
"inference_mode": true,
|
11 |
+
"init_lora_weights": true,
|
12 |
+
"layer_replication": null,
|
13 |
+
"layers_pattern": null,
|
14 |
+
"layers_to_transform": null,
|
15 |
+
"loftq_config": {},
|
16 |
+
"lora_alpha": 64,
|
17 |
+
"lora_dropout": 0.05,
|
18 |
+
"megatron_config": null,
|
19 |
+
"megatron_core": "megatron.core",
|
20 |
+
"modules_to_save": null,
|
21 |
+
"peft_type": "LORA",
|
22 |
+
"r": 32,
|
23 |
+
"rank_pattern": {},
|
24 |
+
"revision": null,
|
25 |
+
"target_modules": [
|
26 |
+
"v_proj",
|
27 |
+
"q_proj"
|
28 |
+
],
|
29 |
+
"task_type": null,
|
30 |
+
"use_dora": false,
|
31 |
+
"use_rslora": false
|
32 |
+
}
|
adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:829fac29b6a03dcfa42cf9b96f1bb7faab8a608e50d70a593b042b6a306ec782
|
3 |
+
size 14176064
|
audio/1.mp3
ADDED
Binary file (116 kB). View file
|
|
audio/10.mp3
ADDED
Binary file (158 kB). View file
|
|
audio/11.mp3
ADDED
Binary file (131 kB). View file
|
|
audio/12.mp3
ADDED
Binary file (121 kB). View file
|
|
audio/13.mp3
ADDED
Binary file (112 kB). View file
|
|
audio/14.mp3
ADDED
Binary file (98.4 kB). View file
|
|
audio/15.mp3
ADDED
Binary file (139 kB). View file
|
|
audio/16.mp3
ADDED
Binary file (105 kB). View file
|
|
audio/17.mp3
ADDED
Binary file (115 kB). View file
|
|
audio/18.mp3
ADDED
Binary file (106 kB). View file
|
|
audio/19.mp3
ADDED
Binary file (135 kB). View file
|
|
audio/2.mp3
ADDED
Binary file (119 kB). View file
|
|
audio/20.mp3
ADDED
Binary file (113 kB). View file
|
|
audio/21.mp3
ADDED
Binary file (131 kB). View file
|
|
audio/22.mp3
ADDED
Binary file (130 kB). View file
|
|
audio/23.mp3
ADDED
Binary file (114 kB). View file
|
|
audio/24.mp3
ADDED
Binary file (140 kB). View file
|
|
audio/25.mp3
ADDED
Binary file (130 kB). View file
|
|
audio/26.mp3
ADDED
Binary file (125 kB). View file
|
|
audio/27.mp3
ADDED
Binary file (137 kB). View file
|
|
audio/28.mp3
ADDED
Binary file (126 kB). View file
|
|
audio/29.mp3
ADDED
Binary file (128 kB). View file
|
|
audio/3.mp3
ADDED
Binary file (120 kB). View file
|
|
audio/30.mp3
ADDED
Binary file (139 kB). View file
|
|
audio/31.mp3
ADDED
Binary file (127 kB). View file
|
|
audio/32.mp3
ADDED
Binary file (135 kB). View file
|
|
audio/33.mp3
ADDED
Binary file (140 kB). View file
|
|
audio/34.mp3
ADDED
Binary file (137 kB). View file
|
|
audio/35.mp3
ADDED
Binary file (124 kB). View file
|
|
audio/36.mp3
ADDED
Binary file (131 kB). View file
|
|
audio/37.mp3
ADDED
Binary file (124 kB). View file
|
|
audio/38.mp3
ADDED
Binary file (137 kB). View file
|
|
audio/39.mp3
ADDED
Binary file (112 kB). View file
|
|
audio/4.mp3
ADDED
Binary file (130 kB). View file
|
|
audio/40.mp3
ADDED
Binary file (127 kB). View file
|
|
audio/41.mp3
ADDED
Binary file (147 kB). View file
|
|
audio/42.mp3
ADDED
Binary file (144 kB). View file
|
|
audio/43.mp3
ADDED
Binary file (115 kB). View file
|
|
audio/44.mp3
ADDED
Binary file (135 kB). View file
|
|
audio/45.mp3
ADDED
Binary file (126 kB). View file
|
|
audio/46.mp3
ADDED
Binary file (117 kB). View file
|
|
audio/47.mp3
ADDED
Binary file (98.4 kB). View file
|
|
audio/48.mp3
ADDED
Binary file (116 kB). View file
|
|
audio/49.mp3
ADDED
Binary file (119 kB). View file
|
|
audio/5.mp3
ADDED
Binary file (154 kB). View file
|
|
audio/50.mp3
ADDED
Binary file (136 kB). View file
|
|
audio/6.mp3
ADDED
Binary file (104 kB). View file
|
|