đ Wav2Vec2-Large-XLSR-53-Mongolian
This model is a fine - tuned version of facebook/wav2vec2-large-xlsr-53 in Mongolian, utilizing the Common Voice dataset. Ensure your speech input is sampled at 16kHz when using this model.
đĻ Information Table
Property |
Details |
Language |
Mongolian |
Datasets |
Common Voice |
Tags |
audio, automatic - speech - recognition, speech, xlsr - fine - tuning - week |
License |
apache - 2.0 |
Model Name |
XLSR Wav2Vec2 Mongolian by Manan Dey |
Task |
Speech Recognition (automatic - speech - recognition) |
Dataset |
Common Voice mn |
Test WER |
43.08 |
đ Quick Start
This model is a fine - tuned version of facebook/wav2vec2-large-xlsr-53 in Mongolian, using the Common Voice dataset. When using this model, ensure that your speech input is sampled at 16kHz.
đģ Usage Examples
Basic Usage
import torch
import torchaudio
from datasets import load_dataset
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
test_dataset = load_dataset("common_voice", "mn", split="test[:2%]")
processor = Wav2Vec2Processor.from_pretrained("manandey/wav2vec2-large-xlsr-mongolian")
model = Wav2Vec2ForCTC.from_pretrained("manandey/wav2vec2-large-xlsr-mongolian")
resampler = torchaudio.transforms.Resample(48_000, 16_000)
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["speech"][:2], 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["sentence"][:2])
đ Documentation
Evaluation
The model can be evaluated as follows on the Mongolian 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", "mn", split="test")
wer = load_metric("wer")
processor = Wav2Vec2Processor.from_pretrained("manandey/wav2vec2-large-xlsr-mongolian")
model = Wav2Vec2ForCTC.from_pretrained("manandey/wav2vec2-large-xlsr-mongolian")
model.to("cuda")
chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\â\%\â\â\īŋŊ\â\â\(\)]'
resampler = torchaudio.transforms.Resample(48_000, 16_000)
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)
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: 43.08%
Training
The Common Voice train
and validation
datasets were used for training.
đ License
This model is licensed under the apache - 2.0 license.