Downloads · 30 days
44
0% of all-time downloads
carlosdanielhernandezmena/wav2vec2-large-xlsr-53-icelandic-ep10-1000h
wav2vec2-large-xlsr-53-icelandic-ep10-1000h is a automatic speech recognition model from carlosdanielhernandezmena. Use it when you need speech turned into text. It is set up for transformers. The card lists the license as cc-by-4.0.
The "wav2vec2-large-xlsr-53-icelandic-ep10-1000h" is an acoustic model suitable for Automatic Speech Recognition in Icelandic. It is the result of fine-tuning the model "facebook/wav2vec2-large-xlsr-53" for 10 epochs…
Downloads · 30 days
44
0% of all-time downloads
All-time downloads
63K
Public
Repo size
2.5 GB
Likes
0
Public
Click a slice to open those files.
.bin1.3 GB · 100%
From the Hugging Face model README
The "wav2vec2-large-xlsr-53-icelandic-ep10-1000h" is an acoustic model suitable for Automatic Speech Recognition in Icelandic. It is the result of fine-tuning the model "facebook/wav2vec2-large-xlsr-53" for 10 epochs with around 1000 hours of Icelandic data developed by the Language and Voice Laboratory. Most of the data is available at public repositories such as LDC, OpenSLR or Clarin.is
The specific list of corpora used to fine-tune the model is:
The fine-tuning process was performed during December (2022) in the servers of the Language and Voice Laboratory (https://lvl.ru.is/) at Reykjavík University (Iceland) by Carlos Daniel Hernández Mena.
import torch
from transformers import Wav2Vec2Processor
from transformers import Wav2Vec2ForCTC
#Load the processor and model.
MODEL_NAME="carlosdanielhernandezmena/wav2vec2-large-xlsr-53-icelandic-ep10-1000h"
processor = Wav2Vec2Processor.from_pretrained(MODEL_NAME)
model = Wav2Vec2ForCTC.from_pretrained(MODEL_NAME)
#Load the dataset
from datasets import load_dataset, load_metric, Audio
ds=load_dataset("language-and-voice-lab/samromur_children", split="test")
#Downsample to 16kHz
ds = ds.cast_column("audio", Audio(sampling_rate=16_000))
#Process the dataset
def prepare_dataset(batch):
audio = batch["audio"]
#Batched output is "un-batched" to ensure mapping is correct
batch["input_values"] = processor(audio["array"], sampling_rate=audio["sampling_rate"]).input_values[0]
with processor.as_target_processor():
batch["labels"] = processor(batch["normalized_text"]).input_ids
return batch
ds = ds.map(prepare_dataset, remove_columns=ds.column_names,num_proc=1)
#Define the evaluation metric
import numpy as np
wer_metric = load_metric("wer")
def compute_metrics(pred):
pred_logits = pred.predictions
pred_ids = np.argmax(pred_logits, axis=-1)
pred.label_ids[pred.label_ids == -100] = processor.tokenizer.pad_token_id
pred_str = processor.batch_decode(pred_ids)
#We do not want to group tokens when computing the metrics
label_str = processor.batch_decode(pred.label_ids, group_tokens=False)
wer = wer_metric.compute(predictions=pred_str, references=label_str)
return {"wer": wer}
#Do the evaluation (with batch_size=1)
model = model.to(torch.device("cuda"))
def map_to_result(batch):
with torch.no_grad():
input_values = torch.tensor(batch["input_values"], device="cuda").unsqueeze(0)
logits = model(input_values).logits
pred_ids = torch.argmax(logits, dim=-1)
batch["pred_str"] = processor.batch_decode(pred_ids)[0]
batch["sentence"] = processor.decode(batch["labels"], group_tokens=False)
return batch
results = ds.map(map_to_result,remove_columns=ds.column_names)
#Compute the overall WER now.
print("Test WER: {:.3f}".format(wer_metric.compute(predictions=results["pred_str"], references=results["sentence"])))
Test Result: 0.094
When publishing results based on these models please refer to:
@misc{mena2022xlrs53icelandic,
title={Acoustic Model in Icelandic: wav2vec2-large-xlsr-53-icelandic-ep10-1000h.},
author={Hernandez Mena, Carlos Daniel},
url={https://huggingface.co/carlosdanielhernandezmena/wav2vec2-large-xlsr-53-icelandic-ep10-1000h},
year={2022}
}
Special thanks to Jón Guðnason, head of the Language and Voice Lab for providing computational power to make this model possible. We also want to thank to the "Language Technology Programme for Icelandic 2019-2023" which is managed and coordinated by Almannarómur, and it is funded by the Icelandic Ministry of Education, Science and Culture.