Loading
Tem um workflow do ComfyUI? Gere uma documentação legível direto no navegador.Torne um workflow legível.Experimentar agora ↗
Fine-tuned facebook/wav2vec2-large-xlsr-53 on the Indonesian Artificial Common Voice dataset. When using this model, make sure that your speech input is sampled at 16kHz.
Fonte do modelo
Descrição da fonte
Fine-tuned facebook/wav2vec2-large-xlsr-53 on the Indonesian Artificial Common Voice dataset. When using this model, make sure that your speech input is sampled at 16kHz.
The model can be used directly (without a language model) as follows:
Fontes
1 fonteVerificado 13 de set.
Artefatos de modelo
2 artefatosTrechos de fonte
2 trechos
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("cahya/wav2vec2-large-xlsr-indonesian")
model = Wav2Vec2ForCTC.from_pretrained("cahya/wav2vec2-large-xlsr-indonesian")
resampler = torchaudio.transforms.Resample(48_000, 16_000)
# 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"])
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"])
The model can be evaluated as follows on the Indonesian test data of Common Voice.
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("cahya/wav2vec2-large-xlsr-indonesian")
model = Wav2Vec2ForCTC.from_pretrained("cahya/wav2vec2-large-xlsr-indonesian")
model.to("cuda")
chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\“\%\‘\'\”\�]'
resampler = torchaudio.transforms.Resample(48_000, 16_000)
# 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"])
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: 51.69 %
The Artificial Common Voice train, validation, and ... datasets were used for training.
The script used for training can be found here (will be available soon)
--- 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 with Artificial Voice by Cahya 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: 51.69 --- # Wav2Vec2-Large-XLSR-Indonesian Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the [Indonesian Artificial Common Voice dataset](https://cloud.uncool.ai/index.php/f/2165181). 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("cahya/wav2vec2-large-xlsr-indonesian") model = Wav2Vec2ForCTC.from_pretrained("cahya/wav2vec2-large-xlsr-indonesian") resampler = torchaudio.transforms.Resample(48_000, 16_000) # 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"]) 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("cahya/wav2vec2-large-xlsr-indonesian") model = Wav2Vec2ForCTC.from_pretr...
Source context: 4 downloads · 0 likes · Pipeline automatic-speech-recognition · Repo takedakimochi/wav2vec2-large-xlsr-indonesian-artificial