commit 00d3b202969924d11931ccce8f08fa02fc747354 Author: ModelHub XC Date: Thu May 28 10:39:18 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: indonesian-nlp/wav2vec2-large-xlsr-indonesian Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..d699711 --- /dev/null +++ b/.gitattributes @@ -0,0 +1,17 @@ +*.bin.* filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tar.gz filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack 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 +*.msgpack filter=lfs diff=lfs merge=lfs -text diff --git a/README.md b/README.md new file mode 100644 index 0000000..8d113c5 --- /dev/null +++ b/README.md @@ -0,0 +1,126 @@ +--- +language: id +datasets: +- common_voice +metrics: +- wer +tags: +- audio +- automatic-speech-recognition +- speech +- xlsr-fine-tuning-week +license: apache-2.0 +model-index: +- name: XLSR Wav2Vec2 Indonesian by Indonesian NLP + results: + - task: + name: Speech Recognition + type: automatic-speech-recognition + dataset: + name: Common Voice id + type: common_voice + args: id + metrics: + - name: Test WER + type: wer + value: 14.29 +--- + +# Wav2Vec2-Large-XLSR-Indonesian + +This is the model for Wav2Vec2-Large-XLSR-Indonesian, a fine-tuned +[facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) +model on the [Indonesian Common Voice dataset](https://huggingface.co/datasets/common_voice). +When using this model, make sure that your speech input is sampled at 16kHz. + +## Usage +The model can be used directly (without a language model) as follows: +```python +import torch +import torchaudio +from datasets import load_dataset +from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor + +test_dataset = load_dataset("common_voice", "id", split="test[:2%]") + +processor = Wav2Vec2Processor.from_pretrained("indonesian-nlp/wav2vec2-large-xlsr-indonesian") +model = Wav2Vec2ForCTC.from_pretrained("indonesian-nlp/wav2vec2-large-xlsr-indonesian") + + +# Preprocessing the datasets. +# We need to read the aduio files as arrays +def speech_file_to_array_fn(batch): + speech_array, sampling_rate = torchaudio.load(batch["path"]) + resampler = torchaudio.transforms.Resample(sampling_rate, 16_000) + batch["speech"] = resampler(speech_array).squeeze().numpy() + return batch + +test_dataset = test_dataset.map(speech_file_to_array_fn) +inputs = processor(test_dataset[:2]["speech"], sampling_rate=16_000, return_tensors="pt", padding=True) + +with torch.no_grad(): + logits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits + +predicted_ids = torch.argmax(logits, dim=-1) + +print("Prediction:", processor.batch_decode(predicted_ids)) +print("Reference:", test_dataset[:2]["sentence"]) +``` + + +## Evaluation + +The model can be evaluated as follows on the Indonesian test data of Common Voice. + +```python +import torch +import torchaudio +from datasets import load_dataset, load_metric +from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor +import re + +test_dataset = load_dataset("common_voice", "id", split="test") +wer = load_metric("wer") + +processor = Wav2Vec2Processor.from_pretrained("indonesian-nlp/wav2vec2-large-xlsr-indonesian") +model = Wav2Vec2ForCTC.from_pretrained("indonesian-nlp/wav2vec2-large-xlsr-indonesian") +model.to("cuda") + +chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\“\%\‘\'\”\�]' + + +# Preprocessing the datasets. +# We need to read the aduio files as arrays +def speech_file_to_array_fn(batch): + batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower() + speech_array, sampling_rate = torchaudio.load(batch["path"]) + resampler = torchaudio.transforms.Resample(sampling_rate, 16_000) + batch["speech"] = resampler(speech_array).squeeze().numpy() + return batch + +test_dataset = test_dataset.map(speech_file_to_array_fn) + +# Preprocessing the datasets. +# We need to read the aduio files as arrays +def evaluate(batch): + inputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True) + + with torch.no_grad(): + logits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits + + pred_ids = torch.argmax(logits, dim=-1) + batch["pred_strings"] = processor.batch_decode(pred_ids) + return batch + +result = test_dataset.map(evaluate, batched=True, batch_size=8) + +print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"]))) +``` + +**Test Result**: 14.29 % + +## Training + +The Common Voice `train`, `validation`, and [synthetic voice datasets](https://cloud.uncool.ai/index.php/s/Kg4C6f5NJGN9ZdR) were used for training. + +The script used for training can be found [here](https://github.com/indonesian-nlp/wav2vec2-indonesian) diff --git a/config.json b/config.json new file mode 100644 index 0000000..4c677a8 --- /dev/null +++ b/config.json @@ -0,0 +1,76 @@ +{ + "_name_or_path": "/root/Work/indonesian-speech-recognition/wav2vec2-large-xlsr-indonesian-artificial/epoch-30", + "activation_dropout": 0.055, + "apply_spec_augment": true, + "architectures": [ + "Wav2Vec2ForCTC" + ], + "attention_dropout": 0.094, + "bos_token_id": 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": true, + "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": 0.04, + "final_dropout": 0.0, + "gradient_checkpointing": true, + "hidden_act": "gelu", + "hidden_dropout": 0.047, + "hidden_size": 1024, + "initializer_range": 0.02, + "intermediate_size": 4096, + "layer_norm_eps": 1e-05, + "layerdrop": 0.041, + "mask_channel_length": 10, + "mask_channel_min_space": 1, + "mask_channel_other": 0.0, + "mask_channel_prob": 0.0, + "mask_channel_selection": "static", + "mask_feature_length": 10, + "mask_feature_prob": 0.0, + "mask_time_length": 10, + "mask_time_min_space": 1, + "mask_time_other": 0.0, + "mask_time_prob": 0.4, + "mask_time_selection": "static", + "model_type": "wav2vec2", + "num_attention_heads": 16, + "num_conv_pos_embedding_groups": 16, + "num_conv_pos_embeddings": 128, + "num_feat_extract_layers": 7, + "num_hidden_layers": 24, + "pad_token_id": 27, + "transformers_version": "4.5.0.dev0", + "vocab_size": 28 +} diff --git a/flax_model.msgpack b/flax_model.msgpack new file mode 100644 index 0000000..34c3208 --- /dev/null +++ b/flax_model.msgpack @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6cce8c0a3fc563c385c607843400be2bf4cae17e368b5fc302ce9e4ae1cdb700 +size 1261885072 diff --git a/preprocessor_config.json b/preprocessor_config.json new file mode 100644 index 0000000..0886a48 --- /dev/null +++ b/preprocessor_config.json @@ -0,0 +1,8 @@ +{ + "do_normalize": true, + "feature_size": 1, + "padding_side": "right", + "padding_value": 0.0, + "return_attention_mask": true, + "sampling_rate": 16000 +} diff --git a/pytorch_model.bin b/pytorch_model.bin new file mode 100644 index 0000000..f738ceb --- /dev/null +++ b/pytorch_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2ac83f343489f702212f47f54db0e5ca203e4ac6cef58b9f4c9d2f5e038fd75e +size 1262048599 diff --git a/special_tokens_map.json b/special_tokens_map.json new file mode 100644 index 0000000..9abf719 --- /dev/null +++ b/special_tokens_map.json @@ -0,0 +1 @@ +{"bos_token": "", "eos_token": "", "unk_token": "[UNK]", "pad_token": "[PAD]"} \ No newline at end of file diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000..a2a8340 --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1 @@ +{"unk_token": "[UNK]", "bos_token": "", "eos_token": "", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|"} \ No newline at end of file diff --git a/vocab.json b/vocab.json new file mode 100644 index 0000000..9f2577e --- /dev/null +++ b/vocab.json @@ -0,0 +1 @@ +{"m": 0, "h": 1, "r": 2, "e": 3, "z": 4, "a": 5, "k": 7, "g": 8, "s": 9, "x": 10, "l": 11, "f": 12, "b": 13, "v": 14, "n": 15, "c": 16, "d": 17, "p": 18, "o": 19, "t": 20, "y": 21, "w": 22, "i": 23, "j": 24, "u": 25, "|": 6, "[UNK]": 26, "[PAD]": 27} \ No newline at end of file