imvladikon/wav2vec2-xls-r-300m-hebrew
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: imvladikon/wav2vec2-xls-r-300m-hebrew
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-05-20 23:00:36 +08:00
commit 4bc6b41f80
19 changed files with 2641 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

28
.gitattributes vendored Normal file
View File

@@ -0,0 +1,28 @@
*.7z filter=lfs diff=lfs merge=lfs -text
*.arrow filter=lfs diff=lfs merge=lfs -text
*.bin filter=lfs diff=lfs merge=lfs -text
*.bin.* filter=lfs diff=lfs merge=lfs -text
*.bz2 filter=lfs diff=lfs merge=lfs -text
*.ftz filter=lfs diff=lfs merge=lfs -text
*.gz filter=lfs diff=lfs merge=lfs -text
*.h5 filter=lfs diff=lfs merge=lfs -text
*.joblib filter=lfs diff=lfs merge=lfs -text
*.lfs.* filter=lfs diff=lfs merge=lfs -text
*.model filter=lfs diff=lfs merge=lfs -text
*.msgpack filter=lfs diff=lfs merge=lfs -text
*.onnx filter=lfs diff=lfs merge=lfs -text
*.ot filter=lfs diff=lfs merge=lfs -text
*.parquet filter=lfs diff=lfs merge=lfs -text
*.pb filter=lfs diff=lfs merge=lfs -text
*.pt filter=lfs diff=lfs merge=lfs -text
*.pth filter=lfs diff=lfs merge=lfs -text
*.rar filter=lfs diff=lfs merge=lfs -text
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
*.tar.* filter=lfs diff=lfs merge=lfs -text
*.tflite filter=lfs diff=lfs merge=lfs -text
*.tgz filter=lfs diff=lfs merge=lfs -text
*.xz filter=lfs diff=lfs merge=lfs -text
*.zip filter=lfs diff=lfs merge=lfs -text
*.zstandard filter=lfs diff=lfs merge=lfs -text
*tfevents* filter=lfs diff=lfs merge=lfs -text
model.safetensors filter=lfs diff=lfs merge=lfs -text

1
.gitignore vendored Normal file
View File

@@ -0,0 +1 @@
checkpoint-*/

251
README.md Normal file
View File

@@ -0,0 +1,251 @@
---
language:
- he
tags:
- automatic-speech-recognition
- generated_from_trainer
- he
- hf-asr-leaderboard
- robust-speech-event
base_model: facebook/wav2vec2-xls-r-300m
model-index:
- name: wav2vec2-xls-r-300m-hebrew
results:
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: Custom Dataset
type: custom
args: he
metrics:
- type: wer
value: 23.18
name: Test WER
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# wav2vec2-xls-r-300m-hebrew
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the private datasets in 2 stages - firstly was fine-tuned on a small dataset with good samples Then the obtained model was fine-tuned on a large dataset with the small good dataset, with various samples from different sources, and with an unlabeled dataset that was weakly labeled using a previously trained model.
Small dataset:
| split |size(gb) | n_samples | duration(hrs)| |
|---|---|---|---|---|
|train|4.19| 20306 | 28 | |
|dev |1.05| 5076 | 7 | |
Large dataset:
| split |size(gb) | n_samples | duration(hrs)| |
|---|---|---|---|---|
|train|12.3| 90777 | 69 | |
|dev |2.39| 20246 | 14* | |
(*weakly labeled data wasn't used in validation set)
After firts training it achieves:
on small dataset
- Loss: 0.5438
- WER: 0.1773
on large dataset
- WER: 0.3811
after second training:
on small dataset
- WER: 0.1697
on large dataset
- Loss: 0.4502
- WER: 0.2318
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
#### First training
The following hyperparameters were used during training:
- learning_rate: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- total_eval_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 100.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:-----:|:-----:|:---------------:|:------:|
| No log | 3.15 | 1000 | 0.5203 | 0.4333 |
| 1.4284 | 6.31 | 2000 | 0.4816 | 0.3951 |
| 1.4284 | 9.46 | 3000 | 0.4315 | 0.3546 |
| 1.283 | 12.62 | 4000 | 0.4278 | 0.3404 |
| 1.283 | 15.77 | 5000 | 0.4090 | 0.3054 |
| 1.1777 | 18.93 | 6000 | 0.3893 | 0.3006 |
| 1.1777 | 22.08 | 7000 | 0.3968 | 0.2857 |
| 1.0994 | 25.24 | 8000 | 0.3892 | 0.2751 |
| 1.0994 | 28.39 | 9000 | 0.4061 | 0.2690 |
| 1.0323 | 31.54 | 10000 | 0.4114 | 0.2507 |
| 1.0323 | 34.7 | 11000 | 0.4021 | 0.2508 |
| 0.9623 | 37.85 | 12000 | 0.4032 | 0.2378 |
| 0.9623 | 41.01 | 13000 | 0.4148 | 0.2374 |
| 0.9077 | 44.16 | 14000 | 0.4350 | 0.2323 |
| 0.9077 | 47.32 | 15000 | 0.4515 | 0.2246 |
| 0.8573 | 50.47 | 16000 | 0.4474 | 0.2180 |
| 0.8573 | 53.63 | 17000 | 0.4649 | 0.2171 |
| 0.8083 | 56.78 | 18000 | 0.4455 | 0.2102 |
| 0.8083 | 59.94 | 19000 | 0.4587 | 0.2092 |
| 0.769 | 63.09 | 20000 | 0.4794 | 0.2012 |
| 0.769 | 66.25 | 21000 | 0.4845 | 0.2007 |
| 0.7308 | 69.4 | 22000 | 0.4937 | 0.2008 |
| 0.7308 | 72.55 | 23000 | 0.4920 | 0.1895 |
| 0.6927 | 75.71 | 24000 | 0.5179 | 0.1911 |
| 0.6927 | 78.86 | 25000 | 0.5202 | 0.1877 |
| 0.6622 | 82.02 | 26000 | 0.5266 | 0.1840 |
| 0.6622 | 85.17 | 27000 | 0.5351 | 0.1854 |
| 0.6315 | 88.33 | 28000 | 0.5373 | 0.1811 |
| 0.6315 | 91.48 | 29000 | 0.5331 | 0.1792 |
| 0.6075 | 94.64 | 30000 | 0.5390 | 0.1779 |
| 0.6075 | 97.79 | 31000 | 0.5459 | 0.1773 |
#### Second training
The following hyperparameters were used during training:
- learning_rate: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- total_eval_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1000
- num_epochs: 60.0
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:-----:|:-----:|:---------------:|:------:|
| No log | 0.7 | 1000 | 0.5371 | 0.3811 |
| 1.3606 | 1.41 | 2000 | 0.5247 | 0.3902 |
| 1.3606 | 2.12 | 3000 | 0.5126 | 0.3859 |
| 1.3671 | 2.82 | 4000 | 0.5062 | 0.3828 |
| 1.3671 | 3.53 | 5000 | 0.4979 | 0.3672 |
| 1.3421 | 4.23 | 6000 | 0.4906 | 0.3816 |
| 1.3421 | 4.94 | 7000 | 0.4784 | 0.3651 |
| 1.328 | 5.64 | 8000 | 0.4810 | 0.3669 |
| 1.328 | 6.35 | 9000 | 0.4747 | 0.3597 |
| 1.3109 | 7.05 | 10000 | 0.4813 | 0.3808 |
| 1.3109 | 7.76 | 11000 | 0.4631 | 0.3561 |
| 1.2873 | 8.46 | 12000 | 0.4603 | 0.3431 |
| 1.2873 | 9.17 | 13000 | 0.4579 | 0.3533 |
| 1.2661 | 9.87 | 14000 | 0.4471 | 0.3365 |
| 1.2661 | 10.58 | 15000 | 0.4584 | 0.3437 |
| 1.249 | 11.28 | 16000 | 0.4461 | 0.3454 |
| 1.249 | 11.99 | 17000 | 0.4482 | 0.3367 |
| 1.2322 | 12.69 | 18000 | 0.4464 | 0.3335 |
| 1.2322 | 13.4 | 19000 | 0.4427 | 0.3454 |
| 1.22 | 14.1 | 20000 | 0.4440 | 0.3395 |
| 1.22 | 14.81 | 21000 | 0.4459 | 0.3378 |
| 1.2044 | 15.51 | 22000 | 0.4406 | 0.3199 |
| 1.2044 | 16.22 | 23000 | 0.4398 | 0.3155 |
| 1.1913 | 16.92 | 24000 | 0.4237 | 0.3150 |
| 1.1913 | 17.63 | 25000 | 0.4287 | 0.3279 |
| 1.1705 | 18.34 | 26000 | 0.4253 | 0.3103 |
| 1.1705 | 19.04 | 27000 | 0.4234 | 0.3098 |
| 1.1564 | 19.75 | 28000 | 0.4174 | 0.3076 |
| 1.1564 | 20.45 | 29000 | 0.4260 | 0.3160 |
| 1.1461 | 21.16 | 30000 | 0.4235 | 0.3036 |
| 1.1461 | 21.86 | 31000 | 0.4309 | 0.3055 |
| 1.1285 | 22.57 | 32000 | 0.4264 | 0.3006 |
| 1.1285 | 23.27 | 33000 | 0.4201 | 0.2880 |
| 1.1135 | 23.98 | 34000 | 0.4131 | 0.2975 |
| 1.1135 | 24.68 | 35000 | 0.4202 | 0.2849 |
| 1.0968 | 25.39 | 36000 | 0.4105 | 0.2888 |
| 1.0968 | 26.09 | 37000 | 0.4210 | 0.2834 |
| 1.087 | 26.8 | 38000 | 0.4123 | 0.2843 |
| 1.087 | 27.5 | 39000 | 0.4216 | 0.2803 |
| 1.0707 | 28.21 | 40000 | 0.4161 | 0.2787 |
| 1.0707 | 28.91 | 41000 | 0.4186 | 0.2740 |
| 1.0575 | 29.62 | 42000 | 0.4118 | 0.2845 |
| 1.0575 | 30.32 | 43000 | 0.4243 | 0.2773 |
| 1.0474 | 31.03 | 44000 | 0.4221 | 0.2707 |
| 1.0474 | 31.73 | 45000 | 0.4138 | 0.2700 |
| 1.0333 | 32.44 | 46000 | 0.4102 | 0.2638 |
| 1.0333 | 33.15 | 47000 | 0.4162 | 0.2650 |
| 1.0191 | 33.85 | 48000 | 0.4155 | 0.2636 |
| 1.0191 | 34.56 | 49000 | 0.4129 | 0.2656 |
| 1.0087 | 35.26 | 50000 | 0.4157 | 0.2632 |
| 1.0087 | 35.97 | 51000 | 0.4090 | 0.2654 |
| 0.9901 | 36.67 | 52000 | 0.4183 | 0.2587 |
| 0.9901 | 37.38 | 53000 | 0.4251 | 0.2648 |
| 0.9795 | 38.08 | 54000 | 0.4229 | 0.2555 |
| 0.9795 | 38.79 | 55000 | 0.4176 | 0.2546 |
| 0.9644 | 39.49 | 56000 | 0.4223 | 0.2513 |
| 0.9644 | 40.2 | 57000 | 0.4244 | 0.2530 |
| 0.9534 | 40.9 | 58000 | 0.4175 | 0.2538 |
| 0.9534 | 41.61 | 59000 | 0.4213 | 0.2505 |
| 0.9397 | 42.31 | 60000 | 0.4275 | 0.2565 |
| 0.9397 | 43.02 | 61000 | 0.4315 | 0.2528 |
| 0.9269 | 43.72 | 62000 | 0.4316 | 0.2501 |
| 0.9269 | 44.43 | 63000 | 0.4247 | 0.2471 |
| 0.9175 | 45.13 | 64000 | 0.4376 | 0.2469 |
| 0.9175 | 45.84 | 65000 | 0.4335 | 0.2450 |
| 0.9026 | 46.54 | 66000 | 0.4336 | 0.2452 |
| 0.9026 | 47.25 | 67000 | 0.4400 | 0.2427 |
| 0.8929 | 47.95 | 68000 | 0.4382 | 0.2429 |
| 0.8929 | 48.66 | 69000 | 0.4361 | 0.2415 |
| 0.8786 | 49.37 | 70000 | 0.4413 | 0.2398 |
| 0.8786 | 50.07 | 71000 | 0.4392 | 0.2415 |
| 0.8714 | 50.78 | 72000 | 0.4345 | 0.2406 |
| 0.8714 | 51.48 | 73000 | 0.4475 | 0.2402 |
| 0.8589 | 52.19 | 74000 | 0.4473 | 0.2374 |
| 0.8589 | 52.89 | 75000 | 0.4457 | 0.2357 |
| 0.8493 | 53.6 | 76000 | 0.4462 | 0.2366 |
| 0.8493 | 54.3 | 77000 | 0.4494 | 0.2356 |
| 0.8395 | 55.01 | 78000 | 0.4472 | 0.2352 |
| 0.8395 | 55.71 | 79000 | 0.4490 | 0.2339 |
| 0.8295 | 56.42 | 80000 | 0.4489 | 0.2318 |
| 0.8295 | 57.12 | 81000 | 0.4469 | 0.2320 |
| 0.8225 | 57.83 | 82000 | 0.4478 | 0.2321 |
| 0.8225 | 58.53 | 83000 | 0.4525 | 0.2326 |
| 0.816 | 59.24 | 84000 | 0.4532 | 0.2316 |
| 0.816 | 59.94 | 85000 | 0.4502 | 0.2318 |
### Framework versions
- Transformers 4.17.0.dev0
- Pytorch 1.10.2+cu102
- Datasets 1.18.2.dev0
- Tokenizers 0.11.0

1
added_tokens.json Normal file
View File

@@ -0,0 +1 @@
{"<s>": 30, "</s>": 31}

14
all_results.json Normal file
View File

@@ -0,0 +1,14 @@
{
"epoch": 60.0,
"eval_loss": 0.4502493739128113,
"eval_runtime": 418.5824,
"eval_samples": 20246,
"eval_samples_per_second": 48.368,
"eval_steps_per_second": 3.024,
"eval_wer": 0.23182961772311134,
"train_loss": 1.0617500136590194,
"train_runtime": 191647.3893,
"train_samples": 90777,
"train_samples_per_second": 28.42,
"train_steps_per_second": 0.444
}

107
config.json Normal file
View File

@@ -0,0 +1,107 @@
{
"_name_or_path": "imvladikon/wav2vec2-xls-r-300m-hebrew",
"activation_dropout": 0.1,
"adapter_kernel_size": 3,
"adapter_stride": 2,
"add_adapter": false,
"apply_spec_augment": true,
"architectures": [
"Wav2Vec2ForCTC"
],
"attention_dropout": 0.0,
"bos_token_id": 1,
"classifier_proj_size": 256,
"codevector_dim": 768,
"contrastive_logits_temperature": 0.1,
"conv_bias": true,
"conv_dim": [
512,
512,
512,
512,
512,
512,
512
],
"conv_kernel": [
10,
3,
3,
3,
3,
2,
2
],
"conv_stride": [
5,
2,
2,
2,
2,
2,
2
],
"ctc_loss_reduction": "mean",
"ctc_zero_infinity": false,
"diversity_loss_weight": 0.1,
"do_stable_layer_norm": true,
"eos_token_id": 2,
"feat_extract_activation": "gelu",
"feat_extract_dropout": 0.0,
"feat_extract_norm": "layer",
"feat_proj_dropout": 0.0,
"feat_quantizer_dropout": 0.0,
"final_dropout": 0.0,
"hidden_act": "gelu",
"hidden_dropout": 0.0,
"hidden_size": 1024,
"initializer_range": 0.02,
"intermediate_size": 4096,
"layer_norm_eps": 1e-05,
"layerdrop": 0.0,
"mask_feature_length": 64,
"mask_feature_min_masks": 0,
"mask_feature_prob": 0.25,
"mask_time_length": 10,
"mask_time_min_masks": 2,
"mask_time_prob": 0.75,
"model_type": "wav2vec2",
"num_adapter_layers": 3,
"num_attention_heads": 16,
"num_codevector_groups": 2,
"num_codevectors_per_group": 320,
"num_conv_pos_embedding_groups": 16,
"num_conv_pos_embeddings": 128,
"num_feat_extract_layers": 7,
"num_hidden_layers": 24,
"num_negatives": 100,
"output_hidden_size": 1024,
"pad_token_id": 29,
"proj_codevector_dim": 768,
"tdnn_dilation": [
1,
2,
3,
1,
1
],
"tdnn_dim": [
512,
512,
512,
512,
1500
],
"tdnn_kernel": [
5,
3,
3,
1,
1
],
"torch_dtype": "float32",
"transformers_version": "4.17.0.dev0",
"use_weighted_layer_sum": false,
"vocab_size": 32,
"xvector_output_dim": 512
}

139
eval.py Normal file
View File

@@ -0,0 +1,139 @@
#!/usr/bin/env python3
import argparse
import re
from typing import Dict
import torch
from datasets import Audio, Dataset, load_dataset, load_metric
from transformers import AutoFeatureExtractor, pipeline
def log_results(result: Dataset, args: Dict[str, str]):
"""DO NOT CHANGE. This function computes and logs the result metrics."""
log_outputs = args.log_outputs
dataset_id = "_".join(args.dataset.split("/") + [args.config, args.split])
# load metric
wer = load_metric("wer")
cer = load_metric("cer")
# compute metrics
wer_result = wer.compute(references=result["target"], predictions=result["prediction"])
cer_result = cer.compute(references=result["target"], predictions=result["prediction"])
# print & log results
result_str = f"WER: {wer_result}\n" f"CER: {cer_result}"
print(result_str)
with open(f"{dataset_id}_eval_results.txt", "w") as f:
f.write(result_str)
# log all results in text file. Possibly interesting for analysis
if log_outputs is not None:
pred_file = f"log_{dataset_id}_predictions.txt"
target_file = f"log_{dataset_id}_targets.txt"
with open(pred_file, "w") as p, open(target_file, "w") as t:
# mapping function to write output
def write_to_file(batch, i):
p.write(f"{i}" + "\n")
p.write(batch["prediction"] + "\n")
t.write(f"{i}" + "\n")
t.write(batch["target"] + "\n")
result.map(write_to_file, with_indices=True)
def remove_niqqud(string: str) -> str:
return ''.join('' if 1456 <= ord(c) <= 1479 else c for c in string)
def normalize_text(text: str) -> str:
"""DO ADAPT FOR YOUR USE CASE. this function normalizes the target text."""
chars_to_ignore_regex = '[,?.!\-\;\:"“%‘”<EFBFBD>—’…–]' # noqa: W605 IMPORTANT: this should correspond to the chars that were ignored during training
text = re.sub(chars_to_ignore_regex, "", text.lower())
text = remove_niqqud(text)
# In addition, we can normalize the target text, e.g. removing new lines characters etc...
# note that order is important here!
token_sequences_to_ignore = ["\n\n", "\n", " ", " "]
for t in token_sequences_to_ignore:
text = " ".join(text.split(t))
return text
def main(args):
# load dataset
dataset = load_dataset(args.dataset, args.config, split=args.split, use_auth_token=True)
# for testing: only process the first two examples as a test
# dataset = dataset.select(range(10))
# load processor
feature_extractor = AutoFeatureExtractor.from_pretrained(args.model_id)
sampling_rate = feature_extractor.sampling_rate
# resample audio
dataset = dataset.cast_column("audio", Audio(sampling_rate=sampling_rate))
# load eval pipeline
if args.device is None:
args.device = 0 if torch.cuda.is_available() else -1
asr = pipeline("automatic-speech-recognition", model=args.model_id, device=args.device)
# map function to decode audio
def map_to_pred(batch):
prediction = asr(
batch["audio"]["array"], chunk_length_s=args.chunk_length_s, stride_length_s=args.stride_length_s
)
batch["prediction"] = prediction["text"]
batch["target"] = normalize_text(batch["sentence"])
return batch
# run inference on all examples
result = dataset.map(map_to_pred, remove_columns=dataset.column_names)
# compute and log_results
# do not change function below
log_results(result, args)
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument(
"--model_id", type=str, required=True, help="Model identifier. Should be loadable with 🤗 Transformers"
)
parser.add_argument(
"--dataset",
type=str,
required=True,
help="Dataset name to evaluate the `model_id`. Should be loadable with 🤗 Datasets",
)
parser.add_argument(
"--config", type=str, required=True, help="Config of the dataset. *E.g.* `'en'` for Common Voice"
)
parser.add_argument("--split", type=str, required=True, help="Split of the dataset. *E.g.* `'test'`")
parser.add_argument(
"--chunk_length_s", type=float, default=None, help="Chunk length in seconds. Defaults to 5 seconds."
)
parser.add_argument(
"--stride_length_s", type=float, default=None, help="Stride of the audio chunks. Defaults to 1 second."
)
parser.add_argument(
"--log_outputs", action="store_true", help="If defined, write outputs to log file for analysis."
)
parser.add_argument(
"--device",
type=int,
default=None,
help="The device to run the pipeline on. -1 for CPU (default), 0 for the first GPU and so on.",
)
args = parser.parse_args()
main(args)

3
model.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:75c4e535a62575ac57cf374dec17fb72fbab973dbd69d0391e47918880cf4e00
size 1261938632

9
preprocessor_config.json Normal file
View File

@@ -0,0 +1,9 @@
{
"do_normalize": true,
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
"feature_size": 1,
"padding_side": "right",
"padding_value": 0.0,
"return_attention_mask": true,
"sampling_rate": 16000
}

3
pytorch_model.bin Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:5f4ee7fbf1fd8b43c413809f06de6b07930c451e471decd5d7219a2704dee64c
size 1262054897

981
run_train.py Normal file
View File

@@ -0,0 +1,981 @@
# !/usr/bin/env python
# coding=utf-8
import functools
import json
import logging
import os
import re
import sys
import warnings
from dataclasses import dataclass, field
from typing import Any, Callable, Dict, List, Optional, Union
import datasets
import numpy as np
import torch
import torchaudio
from datasets import DatasetDict, ReadInstruction, load_dataset, load_metric, concatenate_datasets
try:
import bitsandbytes as bnb
BNB_AVAILABLE = True
except:
BNB_AVAILABLE = False
try:
import wandb
WANDB_AVAILABLE = True
except:
WANDB_AVAILABLE = False
import transformers
from transformers import (
AutoConfig,
AutoFeatureExtractor,
AutoModelForCTC,
AutoTokenizer,
HfArgumentParser,
Trainer,
TrainerCallback, TrainingArguments,
Wav2Vec2Processor,
set_seed,
)
try:
from torch_audiomentations import (
Compose,
AddGaussianNoise,
AddGaussianSNR,
ClippingDistortion,
FrequencyMask,
Gain,
LoudnessNormalization,
Normalize,
PitchShift,
PolarityInversion,
Shift,
TimeMask,
TimeStretch,
)
AUDIOMENTATIONS_AVAILABLE = True
except:
AUDIOMENTATIONS_AVAILABLE = False
try:
from transformers import AutoProcessor
except:
pass
from transformers.trainer_pt_utils import get_parameter_names
from transformers.trainer_utils import get_last_checkpoint, is_main_process
from transformers.utils import check_min_version
from transformers.utils.versions import require_version
# Will error if the minimal version of Transformers is not installed. Remove at your own risks.
check_min_version("4.16.0")
require_version(
"datasets>=1.13.3",
"To fix: pip install -r examples/pytorch/text-classification/requirements.txt",
)
logger = logging.getLogger(__name__)
def list_field(default=None, metadata=None):
return field(default_factory=lambda: default, metadata=metadata)
@dataclass
class ModelArguments:
"""
Arguments pertaining to which model/config/tokenizer we are going to fine-tune from.
"""
model_name_or_path: str = field(
metadata={
"help": "Path to pretrained model or model identifier from huggingface.co/models"
}
)
tokenizer_name_or_path: Optional[str] = field(
default=None,
metadata={
"help": "Path to pretrained tokenizer or tokenizer identifier from huggingface.co/models"
},
)
cache_dir: Optional[str] = field(
default=None,
metadata={
"help": "Where do you want to store the pretrained models downloaded from huggingface.co"
},
)
freeze_feature_encoder: bool = field(
default=True,
metadata={"help": "Whether to freeze the feature encoder layers of the model."},
)
attention_dropout: float = field(
default=0.0,
metadata={"help": "The dropout ratio for the attention probabilities."},
)
activation_dropout: float = field(
default=0.0,
metadata={
"help": "The dropout ratio for activations inside the fully connected layer."
},
)
feat_proj_dropout: float = field(
default=0.0, metadata={"help": "The dropout ratio for the projected features."}
)
hidden_dropout: float = field(
default=0.0,
metadata={
"help": "The dropout probability for all fully connected layers in the embeddings, encoder, and pooler."
},
)
final_dropout: float = field(
default=0.0,
metadata={"help": "The dropout probability for the final projection layer."},
)
mask_time_prob: float = field(
default=0.05,
metadata={
"help": "Probability of each feature vector along the time axis to be chosen as the start of the vector"
"span to be masked. Approximately ``mask_time_prob * sequence_length // mask_time_length`` feature"
"vectors will be masked along the time axis."
},
)
mask_time_length: int = field(
default=10,
metadata={"help": "Length of vector span to mask along the time axis."},
)
mask_feature_prob: float = field(
default=0.0,
metadata={
"help": "Probability of each feature vector along the feature axis to be chosen as the start of the vector"
"span to be masked. Approximately ``mask_feature_prob * sequence_length // mask_feature_length`` feature bins will be masked along the time axis."
},
)
mask_feature_length: int = field(
default=10,
metadata={"help": "Length of vector span to mask along the feature axis."},
)
layerdrop: float = field(default=0.0, metadata={"help": "The LayerDrop probability."})
ctc_loss_reduction: Optional[str] = field(
default="mean",
metadata={
"help": "The way the ctc loss should be reduced. Should be one of 'mean' or 'sum'."
},
)
@dataclass
class DataTrainingArguments:
"""
Arguments pertaining to what data we are going to input our model for training and eval.
Using `HfArgumentParser` we can turn this class
into argparse arguments to be able to specify them on
the command line.
"""
dataset_path: str = field(
default=None,
metadata={
"help": "The configuration name of the dataset to use (via the datasets library)."
}
)
dataset_name: str = field(
default=None,
metadata={
"help": "The configuration name of the dataset to use (via the datasets library)."
},
)
dataset_config_name: str = field(
default=None,
metadata={
"help": "The configuration name of the dataset to use (via the datasets library)."
},
)
train_split_name: str = field(
default="train",
metadata={
"help": "The name of the training data set split to use (via the datasets library). Defaults to 'train'"
},
)
eval_split_name: str = field(
default="validation",
metadata={
"help": "The name of the training data set split to use (via the datasets library). Defaults to 'train'"
},
)
audio_column_name: str = field(
default="audio",
metadata={
"help": "The name of the dataset column containing the audio data. Defaults to 'audio'"
},
)
text_column_name: str = field(
default="text",
metadata={
"help": "The name of the dataset column containing the text data. Defaults to 'text'"
},
)
wav_filesize_column_name: str = field(
default=None,
metadata={
"help": "The name of the dataset column containing the wav filesize. Defaults is None"
},
)
overwrite_cache: bool = field(
default=False,
metadata={"help": "Overwrite the cached preprocessed datasets or not."},
)
preprocessing_num_workers: Optional[int] = field(
default=None,
metadata={"help": "The number of processes to use for the preprocessing."},
)
max_train_samples: Optional[int] = field(
default=None,
metadata={
"help": "For debugging purposes or quicker training, truncate the number of training examples to this "
"value if set."
},
)
max_eval_samples: Optional[int] = field(
default=None,
metadata={
"help": "For debugging purposes or quicker training, truncate the number of validation examples to this "
"value if set."
},
)
chars_to_ignore: Optional[List[str]] = list_field(
default=None,
metadata={"help": "A list of characters to remove from the transcripts."},
)
eval_metrics: List[str] = list_field(
default=["wer"],
metadata={
"help": "A list of metrics the model should be evaluated on. E.g. `'wer cer'`"
},
)
max_duration_in_seconds: float = field(
default=20.0,
metadata={
"help": "Filter audio files that are longer than `max_duration_in_seconds` seconds to 'max_duration_in_seconds`"
},
)
min_duration_in_seconds: float = field(
default=0.0,
metadata={
"help": "Filter audio files that are shorter than `min_duration_in_seconds` seconds"
},
)
preprocessing_only: bool = field(
default=False,
metadata={
"help": "Whether to only do data preprocessing and skip training. "
"This is especially useful when data preprocessing errors out in distributed training due to timeout. "
"In this case, one should run the preprocessing in a non-distributed setup with `preprocessing_only=True` "
"so that the cached datasets can consequently be loaded in distributed training"
},
)
print_samples: bool = field(
default=False,
metadata={
"help": "Print row with validation inference results to stdout after each epoch"
},
)
use_augmentations: bool = field(
default=False,
metadata={
"help": "Use data augmentation during training"
},
)
use_auth_token: str = field(
default="",
metadata={
"help": "If :obj:`True`, will use the token generated when running"
":obj:`transformers-cli login` as HTTP bearer authorization for remote files."
},
)
unk_token: str = field(
default="[UNK]",
metadata={"help": "The unk token for the tokenizer"},
)
pad_token: str = field(
default="[PAD]",
metadata={"help": "The padding token for the tokenizer"},
)
word_delimiter_token: str = field(
default="|",
metadata={"help": "The word delimiter token for the tokenizer"},
)
phoneme_language: Optional[str] = field(
default=None,
metadata={
"help": "The target language that should be used be"
" passed to the tokenizer for tokenization. Note that"
" this is only relevant if the model classifies the"
" input audio to a sequence of phoneme sequences."
},
)
class Augmentator:
def __init__(
self,
apply_gaussian_noise_with_p=0.1,
apply_gain_with_p=0.1,
apply_pitch_shift_with_p=0.1,
apply_time_stretch_with_p=0.1,
augment_proba=0.1,
sample_rate=16_000
):
self.augmentator_fn = None
self.sample_rate = sample_rate
self.augment_proba = augment_proba
all_p = (
apply_gaussian_noise_with_p
+ apply_gain_with_p
+ apply_pitch_shift_with_p
+ apply_time_stretch_with_p
)
if AUDIOMENTATIONS_AVAILABLE and all_p > 0:
self.augmentator_fn = Compose([
TimeStretch(min_rate=0.8, max_rate=1.2, leave_length_unchanged=False,
p=apply_time_stretch_with_p),
PitchShift(min_semitones=-1, max_semitones=1,
p=apply_pitch_shift_with_p),
Gain(min_gain_in_db=-1, max_gain_in_db=1, p=apply_gain_with_p),
AddGaussianNoise(min_amplitude=0.0001, max_amplitude=0.001,
p=apply_gaussian_noise_with_p),
])
def __call__(self, input_values: List[float], *args, **kwargs):
if AUDIOMENTATIONS_AVAILABLE and self.augmentator_fn is not None:
return self.augmentator_fn(samples=np.array(input_values),
sample_rate=self.sample_rate).tolist()
else:
return input_values
@dataclass
class DataCollatorCTCWithPadding:
"""
Data collator that will dynamically pad the inputs received.
Args:
processor (:class:`~transformers.AutoProcessor`)
The processor used for proccessing the data.
padding (:obj:`bool`, :obj:`str` or :class:`~transformers.tokenization_utils_base.PaddingStrategy`, `optional`, defaults to :obj:`True`):
Select a strategy to pad the returned sequences (according to the model's padding side and padding index)
among:
* :obj:`True` or :obj:`'longest'`: Pad to the longest sequence in the batch (or no padding if only a single
sequence if provided).
* :obj:`'max_length'`: Pad to a maximum length specified with the argument :obj:`max_length` or to the
maximum acceptable input length for the model if that argument is not provided.
* :obj:`False` or :obj:`'do_not_pad'` (default): No padding (i.e., can output a batch with sequences of
different lengths).
max_length (:obj:`int`, `optional`):
Maximum length of the ``input_values`` of the returned list and optionally padding length (see above).
max_length_labels (:obj:`int`, `optional`):
Maximum length of the ``labels`` returned list and optionally padding length (see above).
pad_to_multiple_of (:obj:`int`, `optional`):
If set will pad the sequence to a multiple of the provided value.
This is especially useful to enable the use of Tensor Cores on NVIDIA hardware with compute capability >=
7.5 (Volta).
"""
processor: 'AutoProcessor'
padding: Union[bool, str] = "longest"
pad_to_multiple_of: Optional[int] = None
pad_to_multiple_of_labels: Optional[int] = None
augmentator_fn: Optional[Callable] = None
use_augmentations: bool = False
def __call__(
self, features: List[Dict[str, Union[List[int], torch.Tensor]]]
) -> Dict[str, torch.Tensor]:
# split inputs and labels since they have to be of different lenghts and need
# different padding methods
input_features = [
{
"input_values": self.augmentator_fn(feature["input_values"])
if self.use_augmentations
else feature["input_values"]}
for feature in features
]
label_features = [{"input_ids": feature["labels"]} for feature in features]
batch = self.processor.pad(
input_features,
padding=self.padding,
pad_to_multiple_of=self.pad_to_multiple_of,
return_tensors="pt",
)
with self.processor.as_target_processor():
labels_batch = self.processor.pad(
label_features,
padding=self.padding,
pad_to_multiple_of=self.pad_to_multiple_of_labels,
return_tensors="pt",
)
# replace padding with -100 to ignore loss correctly
labels = labels_batch["input_ids"].masked_fill(
labels_batch.attention_mask.ne(1), -100
)
batch["labels"] = labels
return batch
def create_vocabulary_from_data(
datasets: DatasetDict,
text_column_name: str,
train_split_name: str,
word_delimiter_token: Optional[str] = None,
unk_token: Optional[str] = None,
pad_token: Optional[str] = None,
):
# Given training and test labels create vocabulary
def extract_all_chars(batch):
all_text = " ".join(batch[text_column_name])
vocab = list(set(all_text))
return {"vocab": [vocab], "all_text": [all_text]}
print("extract chars")
vocabs = datasets.map(
extract_all_chars,
batched=True,
batch_size=-1,
keep_in_memory=True,
remove_columns=datasets[train_split_name].column_names,
)
# take union of all unique characters in each dataset
print("make vocab_set")
vocab_set = functools.reduce(
lambda vocab_1, vocab_2: set(vocab_1["vocab"][0]) | set(vocab_2["vocab"][0]),
vocabs.values(),
)
vocab_dict = {v: k for k, v in enumerate(sorted(list(vocab_set)))}
# replace white space with delimiter token
if word_delimiter_token is not None:
vocab_dict[word_delimiter_token] = vocab_dict[" "]
del vocab_dict[" "]
# add unk and pad token
if unk_token is not None:
vocab_dict[unk_token] = len(vocab_dict)
if pad_token is not None:
vocab_dict[pad_token] = len(vocab_dict)
return vocab_dict
def speech_file_to_array_fn(batch, audio_column_name, dataset_path=""):
if dataset_path:
dataset_path = os.path.join(dataset_path, batch[audio_column_name])
else:
dataset_path = batch[audio_column_name] if isinstance(batch[audio_column_name],
str) else \
batch[audio_column_name]["path"]
speech_array, sampling_rate = torchaudio.load(dataset_path)
batch[audio_column_name] = {
"array": speech_array[0].numpy(),
"sampling_rate": sampling_rate,
}
return batch
class PrintSamplesPredictionCallback(TrainerCallback):
def __init__(self, processor, eval_dataset):
super(PrintSamplesPredictionCallback, self).__init__()
self.processor = processor
self.eval_dataset = eval_dataset
self.metric_fn = load_metric("wer")
def on_log(
self,
args: Any,
state: Any,
control: Any,
model: Any,
logs: Optional[Any] = None,
**kwargs
):
"""
:param args:
:param state:
:param control:
:param model:
:param logs:
:param kwargs: 'tokenizer', 'optimizer', 'lr_scheduler', 'train_dataloader', 'eval_dataloader'
:return:
"""
if state.is_local_process_zero:
columns = ["id", "prediction", "reference", "audio", "wer"]
data = []
for idx, row in enumerate(self.eval_dataset):
input_dict = self.processor(row["input_values"],
return_tensors="pt", padding=True)
logits = model(input_dict.input_values.to(model.device)).logits
pred_ids = torch.argmax(logits, dim=-1)[0]
prediction = self.processor.decode(pred_ids)
print(f"Prediction: {prediction}")
reference = row['references'].lower()
print(f"\nReference: {reference}")
if WANDB_AVAILABLE:
audio, sample_rate = tuple(row["audio"].values())
audio = wandb.Audio(np.squeeze(audio),
sample_rate=sample_rate)
wer = self.metric_fn.compute(
predictions=[prediction],
references=[reference],
)
data.append([idx, prediction, reference, audio, wer])
if WANDB_AVAILABLE:
table = wandb.Table(data=data, columns=columns)
wandb.run.log({"audio_predictions": table})
def main():
# See all possible arguments in src/transformers/training_args.py
# or by passing the --help flag to this script.
# We now keep distinct sets of args, for a cleaner separation of concerns.
parser = HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments))
if len(sys.argv) == 2 and sys.argv[1].endswith(".json"):
# If we pass only one argument to the script and it's the path to a json file,
# let's parse it to get our arguments.
model_args, data_args, training_args = parser.parse_json_file(
json_file=os.path.abspath(sys.argv[1])
)
else:
model_args, data_args, training_args = parser.parse_args_into_dataclasses()
# Detecting last checkpoint.
last_checkpoint = None
if (
os.path.isdir(training_args.output_dir)
and training_args.do_train
and not training_args.overwrite_output_dir
):
last_checkpoint = get_last_checkpoint(training_args.output_dir)
if last_checkpoint is None and len(os.listdir(training_args.output_dir)) > 0:
raise ValueError(
f"Output directory ({training_args.output_dir}) already exists and is not empty. "
"Use --overwrite_output_dir to overcome."
)
elif last_checkpoint is not None:
logger.info(
f"Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change "
"the `--output_dir` or add `--overwrite_output_dir` to train from scratch."
)
# Setup logging
logging.basicConfig(
format="%(asctime)s - %(levelname)s - %(name)s - %(message)s",
datefmt="%m/%d/%Y %H:%M:%S",
handlers=[logging.StreamHandler(sys.stdout)],
)
logger.setLevel(
logging.INFO if is_main_process(training_args.local_rank) else logging.WARN
)
# Log on each process the small summary:
logger.warning(
f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}"
f"distributed training: {bool(training_args.local_rank != -1)}, 16-bits training: {training_args.fp16}"
)
# Set the verbosity to info of the Transformers logger (on main process only):
if is_main_process(training_args.local_rank):
transformers.utils.logging.set_verbosity_info()
logger.info("Training/evaluation parameters %s", training_args)
# Set seed before initializing model.
set_seed(training_args.seed)
train_split_name = data_args.train_split_name
eval_split_name = data_args.eval_split_name
# 1. First, let's load the dataset
raw_datasets = DatasetDict({
train_split_name: None,
eval_split_name: None,
})
if data_args.dataset_path:
raw_datasets = load_dataset(
"csv",
data_files={
train_split_name: os.path.join(data_args.dataset_path, "train-all.csv"),
eval_split_name: os.path.join(data_args.dataset_path, "eval-all.csv"),
},
)
if training_args.do_train:
if raw_datasets[train_split_name] is None:
raw_datasets[train_split_name] = load_dataset(
data_args.dataset_name,
data_args.dataset_config_name,
split=data_args.train_split_name,
use_auth_token=data_args.use_auth_token,
)
if data_args.audio_column_name not in raw_datasets[train_split_name].column_names:
raise ValueError(
f"--audio_column_name '{data_args.audio_column_name}' not found in dataset. "
"Make sure to set `--audio_column_name` to the correct audio column - one of "
f"{', '.join(raw_datasets['train'].column_names)}."
)
if data_args.text_column_name not in raw_datasets[train_split_name].column_names:
raise ValueError(
f"--text_column_name {data_args.text_column_name} not found in dataset. "
"Make sure to set `--text_column_name` to the correct text column - one of "
f"{', '.join(raw_datasets['train'].column_names)}."
)
if data_args.max_train_samples is not None:
raw_datasets[train_split_name] = raw_datasets[train_split_name].select(
range(data_args.max_train_samples)
)
if data_args.wav_filesize_column_name is not None:
raw_datasets[train_split_name] = raw_datasets[train_split_name].sort(
data_args.wav_filesize_column_name, reverse=True)
if training_args.do_eval:
if raw_datasets[eval_split_name] is None:
raw_datasets[eval_split_name] = load_dataset(
data_args.dataset_name,
data_args.dataset_config_name,
split=data_args.eval_split_name,
use_auth_token=data_args.use_auth_token,
)
if data_args.max_eval_samples is not None:
raw_datasets[eval_split_name] = raw_datasets[eval_split_name].select(
range(data_args.max_eval_samples)
)
if data_args.wav_filesize_column_name is not None:
raw_datasets[eval_split_name] = raw_datasets[eval_split_name].sort(
data_args.wav_filesize_column_name, reverse=True)
# save special tokens for tokenizer
word_delimiter_token = data_args.word_delimiter_token
unk_token = data_args.unk_token
pad_token = data_args.pad_token
# 3. Next, let's load the config as we might need it to create
# the tokenizer
# load config
config = AutoConfig.from_pretrained(
model_args.model_name_or_path,
cache_dir=model_args.cache_dir,
use_auth_token=data_args.use_auth_token,
)
# 4. Next, if no tokenizer file is defined,
# we create the vocabulary of the model by extracting all unique characters from
# the training and evaluation datasets
# We need to make sure that only first rank saves vocabulary
# make sure all processes wait until vocab is created
tokenizer_name_or_path = model_args.tokenizer_name_or_path
tokenizer_kwargs = {}
# 5. Now we can instantiate the feature extractor, tokenizer and model
# Note for distributed training, the .from_pretrained methods guarantee that only
# one local process can concurrently download model & vocab.
with open(os.path.join(tokenizer_name_or_path, "vocab.json"), "r") as fin:
print("loading tokenizer")
print(fin.read())
# load feature_extractor and tokenizer
tokenizer = AutoTokenizer.from_pretrained(
tokenizer_name_or_path,
use_auth_token=data_args.use_auth_token,
**tokenizer_kwargs,
)
feature_extractor = AutoFeatureExtractor.from_pretrained(
model_args.model_name_or_path,
cache_dir=model_args.cache_dir,
use_auth_token=data_args.use_auth_token,
)
# adapt config
config.update(
{
"feat_proj_dropout": model_args.feat_proj_dropout,
"attention_dropout": model_args.attention_dropout,
"hidden_dropout": model_args.hidden_dropout,
"final_dropout": model_args.final_dropout,
"mask_time_prob": model_args.mask_time_prob,
"mask_time_length": model_args.mask_time_length,
"mask_feature_prob": model_args.mask_feature_prob,
"mask_feature_length": model_args.mask_feature_length,
"gradient_checkpointing": training_args.gradient_checkpointing,
"layerdrop": model_args.layerdrop,
"ctc_loss_reduction": model_args.ctc_loss_reduction,
"pad_token_id": tokenizer.pad_token_id,
"vocab_size": len(tokenizer),
"activation_dropout": model_args.activation_dropout,
}
)
# create model
model = AutoModelForCTC.from_pretrained(
model_args.model_name_or_path,
cache_dir=model_args.cache_dir,
config=config,
use_auth_token=data_args.use_auth_token,
)
# freeze encoder
if model_args.freeze_feature_encoder:
model.freeze_feature_encoder()
# 6. Now we preprocess the datasets including loading the audio, resampling and normalization
# Thankfully, `datasets` takes care of automatically loading and resampling the audio,
# so that we just need to set the correct target sampling rate and normalize the input
# via the `feature_extractor`
# make sure that dataset decodes audio with correct sampling rate
# derive max & min input length for sample rate & max duration
audio_column_name = data_args.audio_column_name
num_workers = data_args.preprocessing_num_workers
# `phoneme_language` is only relevant if the model is fine-tuned on phoneme classification
phoneme_language = data_args.phoneme_language
raw_datasets[train_split_name] = raw_datasets[train_split_name].map(
speech_file_to_array_fn,
num_proc=num_workers,
fn_kwargs={"dataset_path": data_args.dataset_path,
"audio_column_name": audio_column_name},
)
raw_datasets[eval_split_name] = raw_datasets[eval_split_name].map(
speech_file_to_array_fn,
num_proc=num_workers,
fn_kwargs={"dataset_path": data_args.dataset_path,
"audio_column_name": audio_column_name},
)
# Preprocessing the datasets.
# We need to read the audio files as arrays and tokenize the targets.
def prepare_dataset(batch):
# load audio
sample = batch[audio_column_name]
inputs = feature_extractor(sample["array"], sampling_rate=sample["sampling_rate"])
batch["input_values"] = inputs.input_values[0]
batch["input_length"] = len(batch["input_values"])
# encode targets
additional_kwargs = {}
if phoneme_language is not None:
additional_kwargs["phonemizer_lang"] = phoneme_language
batch["labels"] = tokenizer(batch[data_args.text_column_name],
**additional_kwargs).input_ids
return batch
print(f"Vectorizing")
with training_args.main_process_first(desc="dataset map preprocessing"):
vectorized_datasets = raw_datasets.map(
prepare_dataset,
remove_columns=next(iter(raw_datasets.values())).column_names,
num_proc=num_workers,
desc="preprocess datasets",
)
# 7. Next, we can prepare the training.
# Let's use word error rate (WER) as our evaluation metric,
# instantiate a data collator and the trainer
# Define evaluation metrics during training, *i.e.* word error rate, character error rate
eval_metrics = {metric: load_metric(metric) for metric in data_args.eval_metrics}
# for large datasets it is advised to run the preprocessing on a
# single machine first with ``args.preprocessing_only`` since there will mostly likely
# be a timeout when running the script in distributed mode.
# In a second step ``args.preprocessing_only`` can then be set to `False` to load the
# cached dataset
if data_args.preprocessing_only:
logger.info(
f"Data preprocessing finished. Files cached at {vectorized_datasets.cache_files}"
)
return
def compute_metrics(pred):
pred_logits = pred.predictions
pred_ids = np.argmax(pred_logits, axis=-1)
pred.label_ids[pred.label_ids == -100] = tokenizer.pad_token_id
pred_str = tokenizer.batch_decode(pred_ids)
# we do not want to group tokens when computing the metrics
label_str = tokenizer.batch_decode(pred.label_ids, group_tokens=False)
metrics = {
k: v.compute(predictions=pred_str, references=label_str)
for k, v in eval_metrics.items()
}
return metrics
# Now save everything to be able to create a single processor later
if is_main_process(training_args.local_rank):
# save feature extractor, tokenizer and config
feature_extractor.save_pretrained(training_args.output_dir)
tokenizer.save_pretrained(training_args.output_dir)
config.save_pretrained(training_args.output_dir)
try:
processor = AutoProcessor.from_pretrained(training_args.output_dir)
except (OSError, KeyError):
warnings.warn(
"Loading a processor from a feature extractor config that does not"
" include a `processor_class` attribute is deprecated and will be removed in v5. Please add the following "
" attribute to your `preprocessor_config.json` file to suppress this warning: "
" `'processor_class': 'Wav2Vec2Processor'`",
FutureWarning,
)
processor = Wav2Vec2Processor.from_pretrained(training_args.output_dir)
# Instantiate custom data collator
data_collator = DataCollatorCTCWithPadding(
processor=processor,
augmentator_fn=Augmentator(),
use_augmentations=data_args.use_augmentations
)
decay_parameters = get_parameter_names(model, [torch.nn.LayerNorm])
decay_parameters = [name for name in decay_parameters if "bias" not in name]
optimizer_grouped_parameters = [
{
"params": [p for n, p in model.named_parameters() if n in decay_parameters],
"weight_decay": training_args.weight_decay,
},
{
"params": [
p for n, p in model.named_parameters() if n not in decay_parameters
],
"weight_decay": 0.0,
},
]
trainer_kwargs = {}
if BNB_AVAILABLE:
optimizer = bnb.optim.Adam8bit(
params=optimizer_grouped_parameters,
betas=(training_args.adam_beta1, training_args.adam_beta2),
eps=training_args.adam_epsilon,
)
trainer_kwargs["optimizers"] = (optimizer, None)
samples_to_log = [
{
**vectorized_datasets[eval_split_name][i],
"references": raw_datasets[eval_split_name][i][data_args.text_column_name],
"audio": raw_datasets[eval_split_name][i][data_args.audio_column_name],
} for i in range(5)
]
trainer = Trainer(
model=model,
data_collator=data_collator,
args=training_args,
compute_metrics=compute_metrics,
train_dataset=vectorized_datasets[
train_split_name] if training_args.do_train else None,
eval_dataset=vectorized_datasets[
eval_split_name] if training_args.do_eval else None,
tokenizer=feature_extractor,
**trainer_kwargs,
callbacks=[PrintSamplesPredictionCallback(
processor=processor,
eval_dataset=samples_to_log)] if data_args.print_samples and training_args.do_eval else None,
)
# 8. Finally, we can start training
# Training
if training_args.do_train:
# use last checkpoint if exist
if last_checkpoint is not None:
checkpoint = last_checkpoint
elif os.path.isdir(model_args.model_name_or_path):
checkpoint = model_args.model_name_or_path
else:
checkpoint = None
train_result = trainer.train(resume_from_checkpoint=checkpoint)
trainer.save_model()
metrics = train_result.metrics
max_train_samples = (
data_args.max_train_samples
if data_args.max_train_samples is not None
else len(vectorized_datasets[train_split_name])
)
metrics["train_samples"] = min(
max_train_samples, len(vectorized_datasets[train_split_name])
)
trainer.log_metrics(train_split_name, metrics)
trainer.save_metrics(train_split_name, metrics)
trainer.save_state()
# Evaluation
results = {}
if training_args.do_eval:
logger.info("*** Evaluate ***")
metrics = trainer.evaluate()
max_eval_samples = (
data_args.max_eval_samples
if data_args.max_eval_samples is not None
else len(vectorized_datasets[eval_split_name])
)
metrics["eval_samples"] = min(max_eval_samples,
len(vectorized_datasets[eval_split_name]))
trainer.log_metrics(eval_split_name, metrics)
trainer.save_metrics(eval_split_name, metrics)
# Write model card and (optionally) push to hub
config_name = (
data_args.dataset_config_name
if data_args.dataset_config_name is not None
else "na"
)
kwargs = {
"language": "he",
"finetuned_from": model_args.model_name_or_path,
"tasks": "speech-recognition",
"tags": ["automatic-speech-recognition", "robust-speech-event", "he"],
"dataset_args": f"Config: {config_name}, Training split: {data_args.train_split_name}, Eval split: {data_args.eval_split_name}",
}
if training_args.push_to_hub:
trainer.push_to_hub(**kwargs)
else:
trainer.create_model_card(**kwargs)
return results
if __name__ == "__main__":
main()

39
run_train.sh Normal file
View File

@@ -0,0 +1,39 @@
export CUDA_VISIBLE_DEVICES="0,1"
python -m torch.distributed.launch --nproc_per_node=2 run_train.py \
--dataset_name="imvladikon/hebrew_speech_???" \
--use_auth_token="???" \
--audio_column_name="audio" \
--text_column_name="sentence" \
--model_name_or_path="imvladikon/wav2vec2-xls-r-300m-hebrew" \
--tokenizer_name_or_path="./wav2vec2-xls-r-300m-hebrew" \
--output_dir="./wav2vec2-xls-r-300m-hebrew" \
--overwrite_output_dir \
--evaluation_strategy="steps" \
--length_column_name="input_length" \
--gradient_checkpointing \
--fp16 \
--group_by_length \
--num_train_epochs="100" \
--per_device_train_batch_size="8" \
--per_device_eval_batch_size="8" \
--gradient_accumulation_steps="4" \
--learning_rate="3e-4" \
--warmup_steps="1000" \
--save_steps="1000" \
--eval_steps="1000" \
--preprocessing_num_workers="$(nproc)" \
--logging_steps="2000" \
--layerdrop="0.0" \
--activation_dropout="0.1" \
--save_total_limit="3" \
--freeze_feature_encoder \
--feat_proj_dropout="0.0" \
--mask_time_prob="0.75" \
--mask_time_length="10" \
--mask_feature_prob="0.25" \
--mask_feature_length="64" \
--do_train --do_eval \
--print_samples \
--use_augmentations \
--push_to_hub

1
special_tokens_map.json Normal file
View File

@@ -0,0 +1 @@
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}

1
tokenizer_config.json Normal file
View File

@@ -0,0 +1 @@
{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "./wav2vec2-xls-r-300m-a-hebrew", "tokenizer_class": "Wav2Vec2CTCTokenizer"}

8
train_results.json Normal file
View File

@@ -0,0 +1,8 @@
{
"epoch": 60.0,
"train_loss": 1.0617500136590194,
"train_runtime": 191647.3893,
"train_samples": 90777,
"train_samples_per_second": 28.42,
"train_steps_per_second": 0.444
}

1042
trainer_state.json Normal file
View File

File diff suppressed because it is too large Load Diff

3
training_args.bin Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:dfeb10e403369a185fe8d387f3aab04ec2ddef0dea3b5d4d22baea6101fdec23
size 3055

9
validation_results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"epoch": 60.0,
"eval_loss": 0.4502493739128113,
"eval_runtime": 418.5824,
"eval_samples": 20246,
"eval_samples_per_second": 48.368,
"eval_steps_per_second": 3.024,
"eval_wer": 0.23182961772311134
}

1
vocab.json Normal file
View File

@@ -0,0 +1 @@
{"א": 1, "ב": 2, "ג": 3, "ד": 4, "ה": 5, "ו": 6, "ז": 7, "ח": 8, "ט": 9, "י": 10, "ך": 11, "כ": 12, "ל": 13, "ם": 14, "מ": 15, "ן": 16, "נ": 17, "ס": 18, "ע": 19, "ף": 20, "פ": 21, "ץ": 22, "צ": 23, "ק": 24, "ר": 25, "ש": 26, "ת": 27, "|": 0, "[UNK]": 28, "[PAD]": 29}