File size: 1,090 Bytes
27f48e1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
---
language:
- ta
metrics:
- wer
library_name: transformers
pipeline_tag: automatic-speech-recognition
---

# Model Card for Model ID

<!-- Provide a quick summary of what the model is/does. -->

This model is the fine-tuned version of Amrrs/wav2vec2-large-xlsr-53-tamil model for Tamil speech data.

## Model Details


## Uses

<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
This model can be used to fine-tune the wav2vec2-large model for Tamil language.

#### Preprocessing [optional]

* Resampling to 16000 Hz
* Log-Mel Spectrogram extraction


#### Training Hyperparameters

training_args = TrainingArguments(
  output_dir="armms-xlsr",
  group_by_length=True,
  per_device_train_batch_size=4,
  gradient_accumulation_steps=3,
  evaluation_strategy="steps",
  num_train_epochs=30,
  gradient_checkpointing=True,
  fp16=True,
  seed = 42,
  save_steps=200,
  eval_steps=200,
  logging_steps=200,
  learning_rate=3e-4,
  warmup_steps=500,
  save_total_limit=2,
  push_to_hub=True,
)