Modelpublic

okuparinen/LIA_300m_simple

sourceHugging Faceapache-2.0updated 1y agoView on Hugging Face

0likes5downloads

Model Card

Simple automatic dialectal transcription of Norwegian

This is a fine-tuned model for automatic dialectal transcription of Norwegian dialect recordings. The model is based on the XLS-R large model. The model has been finetuned on old Norwegian dialect recordings and their corresponding transcriptions. This model outputs simple transcription. The audio recordings are sampled at 16kHz.

Uses

You can use this model for automatic dialectal transcription of Norwegian dialects. Note that this model does not produce standard bokmål or nynorsk text.

How to Get Started with the Model

from transformers import Wav2Vec2Processor, Wav2Vec2ForCTC, Wav2Vec2CTCTokenizer
from datasets import Dataset, Audio
import torch
import pandas as pd

ds = pd.read_csv('CSV_DATA.csv')
ds = ds.dropna(how='any', axis=0)

test = Dataset.from_pandas(skn_test)
test = test.cast_column("AUDIO_PATH_COLUMN", Audio(sampling_rate=16000))

tokenizer = Wav2Vec2CTCTokenizer.from_pretrained("okuparinen/LIA_300m_simple", unk_token="[UNK]", pad_token="[PAD]", word_delimiter_token="|")
model = Wav2Vec2ForCTC.from_pretrained("okuparinen/LIA_300m_simple").to("cuda")
processor = Wav2Vec2Processor.from_pretrained("okuparinen/LIA_300m_simple", tokenizer=tokenizer)

def prepare_dataset(batch):
    audio = batch["AUDIO_PATH"]
    batch["input_values"] = processor(audio["array"], sampling_rate=audio["sampling_rate"]).input_values[0]
    batch["input_length"] = len(batch["input_values"])

    return batch

test_ready = test.map(prepare_dataset, remove_columns=test.column_names)

length = len(test)
predictions = []

for i in range(0, length, 1):
    input_dict = processor(test_ready[i]["input_values"], return_tensors="pt", padding=True)
    logits = model(input_dict.input_values.to("cuda")).logits
    
    pred_ids = torch.argmax(logits, dim=-1)[0]
    
    prediction = processor.decode(pred_ids)
    predictions.append(prediction)

with open("OUTFILE.txt", "w") as f_pred:
    for line in predictions:
        f_pred.write(line + '\n')

Training Data

The training data is an utterance-level version of the LIA Norwegian corpus. The utterance-level version is available at okuparinen/skn.

Evaluation Results

TBA

Citation

BibTeX:

[More Information Needed]