Create README.md

README.md

@@ -0,0 +1,114 @@
+---
+language:
+- hu
+license: apache-2.0
+tags:
+- automatic-speech-recognition
+- mozilla-foundation/common_voice_8_0
+- generated_from_trainer
+- fi
+- robust-speech-event
+- model_for_talk
+datasets:
+- mozilla-foundation/common_voice_8_0
+
+model-index:
+- name: sammy786/wav2vec2-xlsr-finnish
+  results:
+  - task: 
+      name: Automatic Speech Recognition 
+      type: automatic-speech-recognition
+    dataset:
+      name: Common Voice 8
+      type: mozilla-foundation/common_voice_8_0
+      args: fi
+    metrics:
+       - name: Test WER
+         type: wer
+         value: 0.2889
+       - name: Test CER
+         type: cer
+         value: 0.06
+  - task: 
+      name: Automatic Speech Recognition
+      type: automatic-speech-recognition
+    dataset:
+      name: Robust Speech Event - Dev Data
+      type: speech-recognition-community-v2/dev_data
+      args: hu
+    metrics:
+       - name: Test WER
+         type: wer
+         value: 0.2889
+       - name: Test CER
+         type: cer
+         value: 0.06
+---
+# sammy786/wav2vec2-xlsr-finnish
+
+This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - fi dataset.
+It achieves the following results on evaluation set (which is 10 percent of train data set merged with invalidated data, other, dev and validated datasets):
+- Loss: 0.092132
+- Wer: 0.226287
+
+## Model description
+"facebook/wav2vec2-xls-r-300m" was finetuned.
+
+## Intended uses & limitations
+More information needed
+## Training and evaluation data
+Training data - 
+Common voice Finnish train.tsv, dev.tsv, invalidated.tsv and other.tsv
+
+## Training procedure
+For creating the train dataset, all possible datasets were appended and 90-10 split was used. 
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+
+- learning_rate: 0.000095637994662983496
+- train_batch_size: 32
+- eval_batch_size: 16
+- seed: 13
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine_with_restarts
+- lr_scheduler_warmup_steps: 500
+- num_epochs: 80
+- mixed_precision_training: Native AMP
+
+
+### Training results
+
+
+| Step  | Training Loss | Validation Loss | Wer      |
+|:-----:|:-------------:|:---------------:|:--------:|
+| 1000  | 0.593200      | 0.180788        | 0.396133 |
+| 2000  | 0.154900      | 0.112490        | 0.288999 |
+| 3000  | 0.114900      | 0.103121        | 0.267442 |
+| 4000  | 0.093900      | 0.102042        | 0.258950 |
+| 5000  | 0.080700      | 0.100328        | 0.246407 |
+| 6000  | 0.071100      | 0.097676        | 0.247322 |
+| 7000  | 0.064300      | 0.091082        | 0.240267 |
+| 8000  | 0.056300      | 0.094759        | 0.236086 |
+| 9000  | 0.051300      | 0.092355        | 0.234387 |
+| 10000 | 0.046800      | 0.093417        | 0.228377 |
+| 11000 | 0.045000      | 0.092780        | 0.226810 |
+| 12000 | 0.042100      | 0.092132        | 0.226287 |
+
+
+### Framework versions
+- Transformers 4.16.0.dev0
+- Pytorch 1.10.0+cu102
+- Datasets 1.17.1.dev0
+- Tokenizers 0.10.3
+
+#### Evaluation Commands
+
+1. To evaluate on `mozilla-foundation/common_voice_8_0` with split `test`
+
+```bash
+python eval.py --model_id sammy786/wav2vec2-xlsr-finnish --dataset mozilla-foundation/common_voice_8_0 --config fi --split test
+```