gigant/romanian-wav2vec2
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: gigant/romanian-wav2vec2
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-10-30 21:18:46 +08:00
commit 41bd2c7faa
22 changed files with 89479 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-*/

277
README.md Normal file
View File

@@ -0,0 +1,277 @@
---
language:
- ro
license: apache-2.0
tags:
- automatic-speech-recognition
- hf-asr-leaderboard
- robust-speech-event
datasets:
- mozilla-foundation/common_voice_8_0
- gigant/romanian_speech_synthesis_0_8_1
base_model: facebook/wav2vec2-xls-r-300m
model-index:
- name: wav2vec2-ro-300m_01
results:
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: Robust Speech Event
type: speech-recognition-community-v2/dev_data
args: ro
metrics:
- type: wer
value: 46.99
name: Dev WER (without LM)
- type: cer
value: 16.04
name: Dev CER (without LM)
- type: wer
value: 38.63
name: Dev WER (with LM)
- type: cer
value: 14.52
name: Dev CER (with LM)
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: Common Voice
type: mozilla-foundation/common_voice_8_0
args: ro
metrics:
- type: wer
value: 11.73
name: Test WER (without LM)
- type: cer
value: 2.93
name: Test CER (without LM)
- type: wer
value: 7.31
name: Test WER (with LM)
- type: cer
value: 2.17
name: Test CER (with LM)
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: Robust Speech Event - Test Data
type: speech-recognition-community-v2/eval_data
args: ro
metrics:
- type: wer
value: 43.23
name: Test WER
---
You can test this model online with the [**Space for Romanian Speech Recognition**](https://huggingface.co/spaces/gigant/romanian-speech-recognition)
The model ranked **TOP-1** on Romanian Speech Recognition during HuggingFace's Robust Speech Challenge :
* [**The 🤗 Speech Bench**](https://huggingface.co/spaces/huggingface/hf-speech-bench)
* [**Speech Challenge Leaderboard**](https://huggingface.co/spaces/speech-recognition-community-v2/FinalLeaderboard)
# Romanian Wav2Vec2
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the [Common Voice 8.0 - Romanian subset](https://huggingface.co/datasets/mozilla-foundation/common_voice_8_0) dataset, with extra training data from [Romanian Speech Synthesis](https://huggingface.co/datasets/gigant/romanian_speech_synthesis_0_8_1) dataset.
Without the 5-gram Language Model optimization, it achieves the following results on the evaluation set (Common Voice 8.0, Romanian subset, test split):
- Loss: 0.1553
- Wer: 0.1174
- Cer: 0.0294
## Model description
The architecture is based on [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) with a speech recognition CTC head and an added 5-gram language model (using [pyctcdecode](https://github.com/kensho-technologies/pyctcdecode) and [kenlm](https://github.com/kpu/kenlm)) trained on the [Romanian Corpora Parliament](gigant/ro_corpora_parliament_processed) dataset. Those libraries are needed in order for the language model-boosted decoder to work.
## Intended uses & limitations
The model is made for speech recognition in Romanian from audio clips sampled at **16kHz**. The predicted text is lowercased and does not contain any punctuation.
## How to use
Make sure you have installed the correct dependencies for the language model-boosted version to work. You can just run this command to install the `kenlm` and `pyctcdecode` libraries :
```pip install https://github.com/kpu/kenlm/archive/master.zip pyctcdecode```
With the framework `transformers` you can load the model with the following code :
```
from transformers import AutoProcessor, AutoModelForCTC
processor = AutoProcessor.from_pretrained("gigant/romanian-wav2vec2")
model = AutoModelForCTC.from_pretrained("gigant/romanian-wav2vec2")
```
Or, if you want to test the model, you can load the automatic speech recognition pipeline from `transformers` with :
```
from transformers import pipeline
asr = pipeline("automatic-speech-recognition", model="gigant/romanian-wav2vec2")
```
## Example use with the `datasets` library
First, you need to load your data
We will use the [Romanian Speech Synthesis](https://huggingface.co/datasets/gigant/romanian_speech_synthesis_0_8_1) dataset in this example.
```
from datasets import load_dataset
dataset = load_dataset("gigant/romanian_speech_synthesis_0_8_1")
```
You can listen to the samples with the `IPython.display` library :
```
from IPython.display import Audio
i = 0
sample = dataset["train"][i]
Audio(sample["audio"]["array"], rate = sample["audio"]["sampling_rate"])
```
The model is trained to work with audio sampled at 16kHz, so if the sampling rate of the audio in the dataset is different, we will have to resample it.
In the example, the audio is sampled at 48kHz. We can see this by checking `dataset["train"][0]["audio"]["sampling_rate"]`
The following code resample the audio using the `torchaudio` library :
```
import torchaudio
import torch
i = 0
audio = sample["audio"]["array"]
rate = sample["audio"]["sampling_rate"]
resampler = torchaudio.transforms.Resample(rate, 16_000)
audio_16 = resampler(torch.Tensor(audio)).numpy()
```
To listen to the resampled sample :
```
Audio(audio_16, rate=16000)
```
Know you can get the model prediction by running
```
predicted_text = asr(audio_16)
ground_truth = dataset["train"][i]["sentence"]
print(f"Predicted text : {predicted_text}")
print(f"Ground truth : {ground_truth}")
```
## Training and evaluation data
Training data :
- [Common Voice 8.0 - Romanian subset](https://huggingface.co/datasets/mozilla-foundation/common_voice_8_0) : train + validation + other splits
- [Romanian Speech Synthesis](https://huggingface.co/datasets/gigant/romanian_speech_synthesis_0_8_1) : train + test splits
Evaluation data :
- [Common Voice 8.0 - Romanian subset](https://huggingface.co/datasets/mozilla-foundation/common_voice_8_0) : test split
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.003
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 3
- total_train_batch_size: 48
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 50.0
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|
| 2.9272 | 0.78 | 500 | 0.7603 | 0.7734 | 0.2355 |
| 0.6157 | 1.55 | 1000 | 0.4003 | 0.4866 | 0.1247 |
| 0.4452 | 2.33 | 1500 | 0.2960 | 0.3689 | 0.0910 |
| 0.3631 | 3.11 | 2000 | 0.2580 | 0.3205 | 0.0796 |
| 0.3153 | 3.88 | 2500 | 0.2465 | 0.2977 | 0.0747 |
| 0.2795 | 4.66 | 3000 | 0.2274 | 0.2789 | 0.0694 |
| 0.2615 | 5.43 | 3500 | 0.2277 | 0.2685 | 0.0675 |
| 0.2389 | 6.21 | 4000 | 0.2135 | 0.2518 | 0.0627 |
| 0.2229 | 6.99 | 4500 | 0.2054 | 0.2449 | 0.0614 |
| 0.2067 | 7.76 | 5000 | 0.2096 | 0.2378 | 0.0597 |
| 0.1977 | 8.54 | 5500 | 0.2042 | 0.2387 | 0.0600 |
| 0.1896 | 9.32 | 6000 | 0.2110 | 0.2383 | 0.0595 |
| 0.1801 | 10.09 | 6500 | 0.1909 | 0.2165 | 0.0548 |
| 0.174 | 10.87 | 7000 | 0.1883 | 0.2206 | 0.0559 |
| 0.1685 | 11.65 | 7500 | 0.1848 | 0.2097 | 0.0528 |
| 0.1591 | 12.42 | 8000 | 0.1851 | 0.2039 | 0.0514 |
| 0.1537 | 13.2 | 8500 | 0.1881 | 0.2065 | 0.0518 |
| 0.1504 | 13.97 | 9000 | 0.1840 | 0.1972 | 0.0499 |
| 0.145 | 14.75 | 9500 | 0.1845 | 0.2029 | 0.0517 |
| 0.1417 | 15.53 | 10000 | 0.1884 | 0.2003 | 0.0507 |
| 0.1364 | 16.3 | 10500 | 0.2010 | 0.2037 | 0.0517 |
| 0.1331 | 17.08 | 11000 | 0.1838 | 0.1923 | 0.0483 |
| 0.129 | 17.86 | 11500 | 0.1818 | 0.1922 | 0.0489 |
| 0.1198 | 18.63 | 12000 | 0.1760 | 0.1861 | 0.0465 |
| 0.1203 | 19.41 | 12500 | 0.1686 | 0.1839 | 0.0465 |
| 0.1225 | 20.19 | 13000 | 0.1828 | 0.1920 | 0.0479 |
| 0.1145 | 20.96 | 13500 | 0.1673 | 0.1784 | 0.0446 |
| 0.1053 | 21.74 | 14000 | 0.1802 | 0.1810 | 0.0456 |
| 0.1071 | 22.51 | 14500 | 0.1769 | 0.1775 | 0.0444 |
| 0.1053 | 23.29 | 15000 | 0.1920 | 0.1783 | 0.0457 |
| 0.1024 | 24.07 | 15500 | 0.1904 | 0.1775 | 0.0446 |
| 0.0987 | 24.84 | 16000 | 0.1793 | 0.1762 | 0.0446 |
| 0.0949 | 25.62 | 16500 | 0.1801 | 0.1766 | 0.0443 |
| 0.0942 | 26.4 | 17000 | 0.1731 | 0.1659 | 0.0423 |
| 0.0906 | 27.17 | 17500 | 0.1776 | 0.1698 | 0.0424 |
| 0.0861 | 27.95 | 18000 | 0.1716 | 0.1600 | 0.0406 |
| 0.0851 | 28.73 | 18500 | 0.1662 | 0.1630 | 0.0410 |
| 0.0844 | 29.5 | 19000 | 0.1671 | 0.1572 | 0.0393 |
| 0.0792 | 30.28 | 19500 | 0.1768 | 0.1599 | 0.0407 |
| 0.0798 | 31.06 | 20000 | 0.1732 | 0.1558 | 0.0394 |
| 0.0779 | 31.83 | 20500 | 0.1694 | 0.1544 | 0.0388 |
| 0.0718 | 32.61 | 21000 | 0.1709 | 0.1578 | 0.0399 |
| 0.0732 | 33.38 | 21500 | 0.1697 | 0.1523 | 0.0391 |
| 0.0708 | 34.16 | 22000 | 0.1616 | 0.1474 | 0.0375 |
| 0.0678 | 34.94 | 22500 | 0.1698 | 0.1474 | 0.0375 |
| 0.0642 | 35.71 | 23000 | 0.1681 | 0.1459 | 0.0369 |
| 0.0661 | 36.49 | 23500 | 0.1612 | 0.1411 | 0.0357 |
| 0.0629 | 37.27 | 24000 | 0.1662 | 0.1414 | 0.0355 |
| 0.0587 | 38.04 | 24500 | 0.1659 | 0.1408 | 0.0351 |
| 0.0581 | 38.82 | 25000 | 0.1612 | 0.1382 | 0.0352 |
| 0.0556 | 39.6 | 25500 | 0.1647 | 0.1376 | 0.0345 |
| 0.0543 | 40.37 | 26000 | 0.1658 | 0.1335 | 0.0337 |
| 0.052 | 41.15 | 26500 | 0.1716 | 0.1369 | 0.0343 |
| 0.0513 | 41.92 | 27000 | 0.1600 | 0.1317 | 0.0330 |
| 0.0491 | 42.7 | 27500 | 0.1671 | 0.1311 | 0.0328 |
| 0.0463 | 43.48 | 28000 | 0.1613 | 0.1289 | 0.0324 |
| 0.0468 | 44.25 | 28500 | 0.1599 | 0.1260 | 0.0315 |
| 0.0435 | 45.03 | 29000 | 0.1556 | 0.1232 | 0.0308 |
| 0.043 | 45.81 | 29500 | 0.1588 | 0.1240 | 0.0309 |
| 0.0421 | 46.58 | 30000 | 0.1567 | 0.1217 | 0.0308 |
| 0.04 | 47.36 | 30500 | 0.1533 | 0.1198 | 0.0302 |
| 0.0389 | 48.14 | 31000 | 0.1582 | 0.1185 | 0.0297 |
| 0.0387 | 48.91 | 31500 | 0.1576 | 0.1187 | 0.0297 |
| 0.0376 | 49.69 | 32000 | 0.1560 | 0.1182 | 0.0295 |
### Framework versions
- Transformers 4.16.2
- Pytorch 1.10.0+cu111
- Tokenizers 0.11.0
- pyctcdecode 0.3.0
- kenlm

1
added_tokens.json Normal file
View File

@@ -0,0 +1 @@
{"<s>": 34, "</s>": 35}

15
all_results.json Normal file
View File

@@ -0,0 +1,15 @@
{
"epoch": 50.0,
"eval_cer": 0.029360946316590746,
"eval_loss": 0.15534016489982605,
"eval_runtime": 138.9476,
"eval_samples": 3843,
"eval_samples_per_second": 27.658,
"eval_steps_per_second": 3.462,
"eval_wer": 0.11739302551552884,
"train_loss": 0.17004985039278586,
"train_runtime": 77839.5185,
"train_samples": 30943,
"train_samples_per_second": 19.876,
"train_steps_per_second": 0.414
}

1
alphabet.json Normal file
View File

@@ -0,0 +1 @@
{"labels": [" ", "a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z", "\u00e2", "\u00ee", "\u0103", "\u0219", "\u021b", "\u2047", "", "<s>", "</s>"], "is_bpe": false}

107
config.json Normal file
View File

@@ -0,0 +1,107 @@
{
"_name_or_path": "facebook/wav2vec2-xls-r-300m",
"activation_dropout": 0.0,
"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": 0.1,
"mask_feature_length": 10,
"mask_feature_min_masks": 0,
"mask_feature_prob": 0.0,
"mask_time_length": 10,
"mask_time_min_masks": 2,
"mask_time_prob": 0.05,
"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": 33,
"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": "4.16.0.dev0",
"use_weighted_layer_sum": false,
"vocab_size": 36,
"xvector_output_dim": 512
}

140
eval.py Normal file
View File

@@ -0,0 +1,140 @@
#!/usr/bin/env python3
import argparse
import re
from typing import Dict
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 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())
# 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))
#Replace some characters that are not in the Romanian alphabet but are found in the dataset
text = re.sub('[á]', 'a', text)
text = re.sub('[é]', 'e', text)
text = re.sub('[í]', 'i', text)
text = re.sub('[ò]', 'o', text)
text = re.sub('[ü]', 'u', text)
text = re.sub('[ć]', 'c', text)
text = re.sub('[đ]', 'd', text)
text = re.sub('[č]', 'c', text)
text = re.sub('[š]', 's', text)
text = re.sub('[ş]', 'ș', text) #Replace the s-cedilla that is sometimes (wrongly) used instead of the s-comma
text = re.sub('[ţ]', 'ț', text) #Replace the t-cedilla that is sometimes (wrongly) used instead of the t-comma
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
asr = pipeline("automatic-speech-recognition", model=args.model_id)
# 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."
)
args = parser.parse_args()
main(args)

7
eval.sh Normal file
View File

@@ -0,0 +1,7 @@
python eval.py \
--model_id="gigant/romanian-wav2vec2" \
--dataset="speech-recognition-community-v2/dev_data" \
--config="ro" \
--split="validation" \
--chunk_length_s 10 \
--stride_length_s 1 \

10
eval_results.json Normal file
View File

@@ -0,0 +1,10 @@
{
"epoch": 50.0,
"eval_cer": 0.029360946316590746,
"eval_loss": 0.15534016489982605,
"eval_runtime": 138.9476,
"eval_samples": 3843,
"eval_samples_per_second": 27.658,
"eval_steps_per_second": 3.462,
"eval_wer": 0.11739302551552884
}

3
language_model/5gram.bin Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:b03301f34e511db37a2b00ea1bd61eb4db755478bea45c74498f9f97d3762dff
size 491769893

1
language_model/attrs.json Normal file
View File

@@ -0,0 +1 @@
{"alpha": 0.5, "beta": 1.5, "unk_score_offset": -10.0, "score_boundary": true}

87809
language_model/unigrams.txt Normal file
View File

File diff suppressed because it is too large Load Diff

3
model.safetensors Normal file
View File

@@ -0,0 +1,3 @@
version https://git-lfs.github.com/spec/v1
oid sha256:a42ddf6318dfb62d1375887f03cf8d725d7c91e257f8b86487384d4e8d98a800
size 1261955032

10
preprocessor_config.json Normal file
View File

@@ -0,0 +1,10 @@
{
"do_normalize": true,
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
"feature_size": 1,
"padding_side": "right",
"padding_value": 0,
"processor_class": "Wav2Vec2ProcessorWithLM",
"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:d0b801c7a4fa42854484db39077806d1d0997f8bb7dab06721d2d7f5a6e90ccc
size 1262071281

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}]}

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": "gigant/wav2vec2-ro-300m_01", "tokenizer_class": "Wav2Vec2CTCTokenizer", "processor_class": "Wav2Vec2ProcessorWithLM"}

8
train_results.json Normal file
View File

@@ -0,0 +1,8 @@
{
"epoch": 50.0,
"train_loss": 0.17004985039278586,
"train_runtime": 77839.5185,
"train_samples": 30943,
"train_samples_per_second": 19.876,
"train_steps_per_second": 0.414
}

1049
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:cd7ffa627b29ea27b82a7931cc22a9fb4a091533e719e9390368c2eaf451a89f
size 3055

1
vocab.json Normal file
View File

@@ -0,0 +1 @@
{"a": 1, "b": 2, "c": 3, "d": 4, "e": 5, "f": 6, "g": 7, "h": 8, "i": 9, "j": 10, "k": 11, "l": 12, "m": 13, "n": 14, "o": 15, "p": 16, "q": 17, "r": 18, "s": 19, "t": 20, "u": 21, "v": 22, "w": 23, "x": 24, "y": 25, "z": 26, "â": 27, "î": 28, "ă": 29, "ș": 30, "ț": 31, "|": 0, "[UNK]": 32, "[PAD]": 33}