Update readme for more info

README.md

@@ -26,21 +26,21 @@ model-index:
     metrics:
        - name: Test WER
          type: wer
-         value: 32.09
+         value: 32.18
        - name: Test CER
          type: cer
-         value: 8.23
+         value: 8.27
         
 ---
 
-# Wav2Vec2-Large-XLSR-53 Persian
+# Wav2Vec2-Large-XLSR-53-Persian
 
 Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) in Persian (Farsi) using [Common Voice](https://huggingface.co/datasets/common_voice). When using this model, make sure that your speech input is sampled at 16kHz.
 
-## How To Use
+## Usage
 The model can be used directly (without a language model) as follows:
 
-### Requirements
+**Requirements**
 ```bash
 # requirement packages
 !pip install git+https://github.com/huggingface/datasets.git
@@ -51,23 +51,28 @@ The model can be used directly (without a language model) as follows:
 !pip install hazm
 ```
 
-### Preprocessing
 
+**Prediction**
 ```python
-# preprocessing the datasets.
-# normalizing the texts
+import librosa
+import torch
+import torchaudio
+from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
+from datasets import load_dataset
 
+import numpy as np
 import hazm
 import re
 import string
 
+import IPython.display as ipd
 
 _normalizer = hazm.Normalizer()
-chars_to_ignore = list(set([
+chars_to_ignore = [
     ",", "?", ".", "!", "-", ";", ":", '""', "%", "'", '"', "�",
     "#", "!", "؟", "?", "«", "»", "ء", "،", "(", ")", "؛", "'ٔ", "٬",'ٔ', ",", "?", 
     ".", "!", "-", ";", ":",'"',"“", "%", "‘", "”", "�", "–", "…", "_", "”", '“', '„'
-]))
+]
 
 # In case of farsi
 chars_to_ignore = chars_to_ignore + list(string.ascii_lowercase + string.digits)
@@ -100,37 +105,6 @@ def normalizer(batch, chars_to_ignore, chars_to_mapping):
 
     batch["sentence"] = text
     return batch
-```
-
-### Loading The Data
-
-```python
-from datasets import load_dataset
-
-dataset = load_dataset("common_voice", "fa", split="test[:1%]")
-print(dataset)
-```
-
-**Output:**
-```text
->>> 
-Dataset({
-    features: ['client_id', 'path', 'sentence', 'up_votes', 'down_votes', 'age', 'gender', 'accent', 'locale', 'segment'],
-    num_rows: 52
-})
-```
-
-### Model
-
-```python
-import librosa
-import torch
-import torchaudio
-from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
-
-device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-processor = Wav2Vec2Processor.from_pretrained("m3hrdadfi/wav2vec2-large-xlsr-persian")
-model = Wav2Vec2ForCTC.from_pretrained("m3hrdadfi/wav2vec2-large-xlsr-persian").to(device)
 
 
 def speech_file_to_array_fn(batch):
@@ -155,13 +129,11 @@ def predict(batch):
 
     batch["predicted"] = processor.batch_decode(pred_ids)[0]
     return batch
-```
 
-## Prediction
-
-```python
-import IPython.display as ipd
 
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+processor = Wav2Vec2Processor.from_pretrained("m3hrdadfi/wav2vec2-large-xlsr-persian")
+model = Wav2Vec2ForCTC.from_pretrained("m3hrdadfi/wav2vec2-large-xlsr-persian").to(device)
 
 dataset = load_dataset("common_voice", "fa", split="test[:1%]")
 dataset = dataset.map(
@@ -246,13 +218,91 @@ predicted: من سفر کردم را دوست دارم
 
 ## Evaluation
 
+The model can be evaluated as follows on the Persian (Farsi) test data of Common Voice.
+
 ```bash
-mkdir cer
-wget -O cer/cer.py https://huggingface.co/ctl/wav2vec2-large-xlsr-cantonese/raw/main/cer.py
+!mkdir cer
+!wget -O cer/cer.py https://huggingface.co/ctl/wav2vec2-large-xlsr-cantonese/raw/main/cer.py
 ```
 
 ```python
-from datasets import load_metric
+import librosa
+import torch
+import torchaudio
+from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
+from datasets import load_dataset, load_metric
+
+import hazm
+import re
+import string
+
+_normalizer = hazm.Normalizer()
+chars_to_ignore = [
+    ",", "?", ".", "!", "-", ";", ":", '""', "%", "'", '"', "�",
+    "#", "!", "؟", "?", "«", "»", "ء", "،", "(", ")", "؛", "'ٔ", "٬",'ٔ', ",", "?", 
+    ".", "!", "-", ";", ":",'"',"“", "%", "‘", "”", "�", "–", "…", "_", "”", '“', '„'
+]
+
+# In case of farsi
+# chars_to_ignore = chars_to_ignore + list(string.ascii_lowercase + string.digits)
+
+chars_to_mapping = {
+'ك': 'ک', 'دِ': 'د', 'بِ': 'ب', 'زِ': 'ز', 'ذِ': 'ذ', 'شِ': 'ش', 'سِ': 'س', 'ى': 'ی',
+'ي': 'ی', 'أ': 'ا', 'ؤ': 'و', "ے": "ی", "ۀ": "ه", "ﭘ": "پ", "ﮐ": "ک", "ﯽ": "ی",
+"ﺎ": "ا", "ﺑ": "ب", "ﺘ": "ت", "ﺧ": "خ", "ﺩ": "د", "ﺱ": "س", "ﻀ": "ض", "ﻌ": "ع",
+"ﻟ": "ل", "ﻡ": "م", "ﻢ": "م", "ﻪ": "ه", "ﻮ": "و", "ئ": "ی", 'ﺍ': "ا", 'ة': "ه",
+'ﯾ': "ی", 'ﯿ': "ی", 'ﺒ': "ب", 'ﺖ': "ت", 'ﺪ': "د", 'ﺮ': "ر", 'ﺴ': "س", 'ﺷ': "ش",
+'ﺸ': "ش", 'ﻋ': "ع", 'ﻤ': "م", 'ﻥ': "ن", 'ﻧ': "ن", 'ﻭ': "و", 'ﺭ': "ر", "ﮔ": "گ",
+"\u200c": " ", "\u200d": " ", "\u200e": " ", "\u200f": " ", "\ufeff": " ",
+}
+
+def multiple_replace(text, chars_to_mapping):
+    pattern = "|".join(map(re.escape, chars_to_mapping.keys()))
+    return re.sub(pattern, lambda m: chars_to_mapping[m.group()], str(text))
+
+def remove_special_characters(text, chars_to_ignore_regex):
+    text = re.sub(chars_to_ignore_regex, '', text).lower() + " "
+    return text
+
+def normalizer(batch, chars_to_ignore, chars_to_mapping):
+    chars_to_ignore_regex = f"""[{"".join(chars_to_ignore)}]"""
+    text = batch["sentence"].lower().strip()
+    
+    text = _normalizer.normalize(text)
+    text = multiple_replace(text, chars_to_mapping)
+    text = remove_special_characters(text, chars_to_ignore_regex)
+
+    batch["sentence"] = text
+    return batch
+
+
+def speech_file_to_array_fn(batch):
+    speech_array, sampling_rate = torchaudio.load(batch["path"])
+    speech_array = speech_array.squeeze().numpy()
+    speech_array = librosa.resample(np.asarray(speech_array), sampling_rate, 16_000)
+
+    batch["speech"] = speech_array
+    return batch
+
+
+def predict(batch):
+    features = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
+
+    input_values = features.input_values.to(device)
+    attention_mask = features.attention_mask.to(device)
+
+    with torch.no_grad():
+        logits = model(input_values, attention_mask=attention_mask).logits 
+        
+    pred_ids = torch.argmax(logits, dim=-1)
+
+    batch["predicted"] = processor.batch_decode(pred_ids)[0]
+    return batch
+
+
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+processor = Wav2Vec2Processor.from_pretrained("m3hrdadfi/wav2vec2-large-xlsr-persian")
+model = Wav2Vec2ForCTC.from_pretrained("m3hrdadfi/wav2vec2-large-xlsr-persian").to(device)
 
 dataset = load_dataset("common_voice", "fa", split="test")
 dataset = dataset.map(
@@ -270,11 +320,9 @@ print("WER: {:.2f}".format(100 * wer.compute(predictions=result["predicted"], re
 print("CER: {:.2f}".format(100 * cer.compute(predictions=result["predicted"], references=result["sentence"])))
 ```
 
-**Output:** 
-```text
-WER: 32.09%
-CER: 8.23%
-```
+**Test Result:** 
+- WER: 32.18%
+- CER: 8.27%
 
 
 ## Training