vasilis commited on
Commit
6009cf8
1 Parent(s): 6833db1

Add model files

Browse files
README.md ADDED
@@ -0,0 +1,139 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: et
3
+ datasets:
4
+ - common_voice
5
+ - NST Estonian ASR Database
6
+ metrics:
7
+ - wer
8
+ - cer
9
+ tags:
10
+ - audio
11
+ - automatic-speech-recognition
12
+ - speech
13
+ - xlsr-fine-tuning-week
14
+ license: apache-2.0
15
+ model-index:
16
+ - name: XLSR Wav2Vec2 Large 53 - Estonian by Vasilis
17
+ results:
18
+ - task:
19
+ name: Speech Recognition
20
+ type: automatic-speech-recognition
21
+ dataset:
22
+ name: Common Voice et
23
+ type: common_voice
24
+ args: et
25
+ metrics:
26
+ - name: Test WER
27
+ type: wer
28
+ value: 30.658320
29
+ - name: Test CER
30
+ type: cer
31
+ value: 5.261490
32
+ ---
33
+
34
+ # Wav2Vec2-Large-XLSR-53-Estonian
35
+
36
+ Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on Estonian using the [Common Voice](https://huggingface.co/datasets/common_voice).
37
+ When using this model, make sure that your speech input is sampled at 16kHz.
38
+
39
+ ## Usage
40
+
41
+ The model can be used directly (without a language model) as follows:
42
+
43
+ ```python
44
+ import torch
45
+ import torchaudio
46
+ from datasets import load_dataset
47
+ from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
48
+
49
+ test_dataset = load_dataset("common_voice", "et", split="test[:2%]") #TODO: replace {lang_id} in your language code here. Make sure the code is one of the *ISO codes* of [this](https://huggingface.co/languages) site.
50
+
51
+ processor = Wav2Vec2Processor.from_pretrained("vasilis/wav2vec2-large-xlsr-53-Estonian") #TODO: replace {model_id} with your model id. The model id consists of {your_username}/{your_modelname}, *e.g.* `elgeish/wav2vec2-large-xlsr-53-arabic`
52
+ model = Wav2Vec2ForCTC.from_pretrained("vasilis/wav2vec2-large-xlsr-53-Estonian") #TODO: replace {model_id} with your model id. The model id consists of {your_username}/{your_modelname}, *e.g.* `elgeish/wav2vec2-large-xlsr-53-arabic`
53
+
54
+ resampler = torchaudio.transforms.Resample(48_000, 16_000)
55
+
56
+ # Preprocessing the datasets.
57
+ # We need to read the aduio files as arrays
58
+ def speech_file_to_array_fn(batch):
59
+ speech_array, sampling_rate = torchaudio.load(batch["path"])
60
+ batch["speech"] = resampler(speech_array).squeeze().numpy()
61
+ return batch
62
+
63
+ test_dataset = test_dataset.map(speech_file_to_array_fn)
64
+ inputs = processor(test_dataset["speech"][:2], sampling_rate=16_000, return_tensors="pt", padding=True)
65
+
66
+ with torch.no_grad():
67
+ logits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits
68
+
69
+ predicted_ids = torch.argmax(logits, dim=-1)
70
+
71
+ print("Prediction:", processor.batch_decode(predicted_ids))
72
+ print("Reference:", test_dataset["sentence"][:2])
73
+ ```
74
+
75
+
76
+ ## Evaluation
77
+
78
+ The model can be evaluated as follows on the Estonian test data of Common Voice.
79
+
80
+
81
+ ```python
82
+ import torch
83
+ import torchaudio
84
+ from datasets import load_dataset, load_metric
85
+ from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
86
+ import re
87
+
88
+ test_dataset = load_dataset("common_voice", "et", split="test")
89
+ wer = load_metric("wer")
90
+
91
+ processor = Wav2Vec2Processor.from_pretrained("vasilis/wav2vec2-large-xlsr-53-Estonian")
92
+ model = Wav2Vec2ForCTC.from_pretrained("vasilis/wav2vec2-large-xlsr-53-Estonian")
93
+ model.to("cuda")
94
+
95
+ chars_to_ignore_regex = "[\,\?\.\!\-\;\:\"\“\%\‘\”\�\']" # TODO: adapt this list to include all special characters you removed from the data
96
+
97
+ resampler = {
98
+ 48_000: torchaudio.transforms.Resample(48_000, 16_000),
99
+ 44100: torchaudio.transforms.Resample(44100, 16_000),
100
+ 32000: torchaudio.transforms.Resample(32000, 16_000)
101
+ }
102
+
103
+ # Preprocessing the datasets.
104
+ # We need to read the aduio files as arrays
105
+ def speech_file_to_array_fn(batch):
106
+ batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()
107
+ speech_array, sampling_rate = torchaudio.load(batch["path"])
108
+ batch["speech"] = resampler[sampling_rate](speech_array).squeeze().numpy()
109
+ return batch
110
+
111
+ test_dataset = test_dataset.map(speech_file_to_array_fn)
112
+
113
+ # Preprocessing the datasets.
114
+ # We need to read the aduio files as arrays
115
+ def evaluate(batch):
116
+ inputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
117
+ with torch.no_grad():
118
+ logits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits
119
+ pred_ids = torch.argmax(logits, dim=-1)
120
+ batch["pred_strings"] = processor.batch_decode(pred_ids)
121
+ return batch
122
+
123
+ result = test_dataset.map(evaluate, batched=True, batch_size=8)
124
+
125
+ print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
126
+ print("CER: {:2f}".format(100 * wer.compute(predictions=[" ".join(list(entry)) for entry in result["pred_strings"]], references=[" ".join(list(entry)) for entry in result["sentence"]])))
127
+
128
+ ```
129
+
130
+ **Test Result**: 30.658320 %
131
+
132
+ ## Training
133
+
134
+ Common voice `train` and `validation` sets were used for finetuning
135
+ for 20000 steps (approx. 116 epochs). Both the `feature extractor` (`Wav2Vec2FeatureExtractor`) and
136
+ `feature projection` (`Wav2Vec2FeatureProjection`) layer were frozen. Only the `encoder` layer (`Wav2Vec2EncoderStableLayerNorm`) was finetuned.
137
+
138
+
139
+
config.json ADDED
@@ -0,0 +1,76 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/wav2vec2-large-xlsr-53",
3
+ "activation_dropout": 0.07,
4
+ "apply_spec_augment": true,
5
+ "architectures": [
6
+ "Wav2Vec2Ablation"
7
+ ],
8
+ "attention_dropout": 0.2,
9
+ "bos_token_id": 1,
10
+ "conv_bias": true,
11
+ "conv_dim": [
12
+ 512,
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512
19
+ ],
20
+ "conv_kernel": [
21
+ 10,
22
+ 3,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 2,
27
+ 2
28
+ ],
29
+ "conv_stride": [
30
+ 5,
31
+ 2,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2
37
+ ],
38
+ "ctc_loss_reduction": "mean",
39
+ "ctc_zero_infinity": true,
40
+ "do_stable_layer_norm": true,
41
+ "eos_token_id": 2,
42
+ "feat_extract_activation": "gelu",
43
+ "feat_extract_dropout": 0.0,
44
+ "feat_extract_norm": "layer",
45
+ "feat_proj_dropout": 0.0,
46
+ "final_dropout": 0.0,
47
+ "gradient_checkpointing": true,
48
+ "hidden_act": "gelu",
49
+ "hidden_dropout": 0.05,
50
+ "hidden_size": 1024,
51
+ "initializer_range": 0.02,
52
+ "intermediate_size": 4096,
53
+ "layer_norm_eps": 1e-05,
54
+ "layerdrop": 0.04,
55
+ "mask_channel_length": 10,
56
+ "mask_channel_min_space": 1,
57
+ "mask_channel_other": 0.0,
58
+ "mask_channel_prob": 0.0,
59
+ "mask_channel_selection": "static",
60
+ "mask_feature_length": 10,
61
+ "mask_feature_prob": 0.0,
62
+ "mask_time_length": 10,
63
+ "mask_time_min_space": 1,
64
+ "mask_time_other": 0.0,
65
+ "mask_time_prob": 0.09,
66
+ "mask_time_selection": "static",
67
+ "model_type": "wav2vec2",
68
+ "num_attention_heads": 16,
69
+ "num_conv_pos_embedding_groups": 16,
70
+ "num_conv_pos_embeddings": 128,
71
+ "num_feat_extract_layers": 7,
72
+ "num_hidden_layers": 24,
73
+ "pad_token_id": 35,
74
+ "transformers_version": "4.5.0.dev0",
75
+ "vocab_size": 36
76
+ }
preprocessor_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_size": 1,
4
+ "padding_side": "right",
5
+ "padding_value": 0.0,
6
+ "return_attention_mask": true,
7
+ "sampling_rate": 16000
8
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3a0948e194390a56addc5927ecdfb7a379bf25838cbda9c111f89bf5c342ffb6
3
+ size 1262081431
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]"}
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|"}
vocab.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"r": 0, "ä": 1, "z": 2, "̇": 3, "m": 4, "ö": 5, "w": 6, "õ": 7, "y": 8, "e": 9, "o": 10, "l": 11, "d": 12, "b": 13, "f": 15, "n": 16, "s": 17, "q": 18, "p": 19, "a": 20, "c": 21, "u": 22, "j": 23, "š": 24, "v": 25, "x": 26, "ž": 27, "i": 28, "k": 29, "h": 30, "t": 31, "g": 32, "ü": 33, "|": 14, "[UNK]": 34, "[PAD]": 35}