whisper-large-icelandic-62640-steps-967h

The "whisper-large-icelandic-62640-steps-967h" is an acoustic model suitable for Automatic Speech Recognition in Icelandic. It is the result of fine-tuning the model openai/whisper-large for 62,640 steps with 967 hours of Icelandic data collected by the Language and Voice Laboratory through the platform Samrómur.

The specific data that was used to fine-tune the model is the corpus Samrómur Milljón, which is the result of the automatic verification of 1 million of recordings comming from the corpus "Samromur Unverified 22.07". It has to be pointed out that this model was trained with different data than our previous model whisper-large-icelandic-30k-steps-1000h.

The fine-tuning process was performed during June (2023) in the servers of the Language and Voice Laboratory (https://lvl.ru.is/) at Reykjavík University (Iceland) by Carlos Daniel Hernández Mena. Some details of the creation of this model are reported in the paper Samrómur Milljón: An ASR Corpus of One Million Verified Read Prompts in Icelandic.

Evaluation

import torch
from transformers import WhisperForConditionalGeneration, WhisperProcessor

#Load the processor and model.
MODEL_NAME="language-and-voice-lab/whisper-large-icelandic-62640-steps-967h"
processor = WhisperProcessor.from_pretrained(MODEL_NAME)
model = WhisperForConditionalGeneration.from_pretrained(MODEL_NAME).to("cuda")

#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 map_to_pred(batch):
	audio = batch["audio"]
	input_features = processor(audio["array"], sampling_rate=audio["sampling_rate"], return_tensors="pt").input_features
	batch["reference"] = processor.tokenizer._normalize(batch['normalized_text'])

	with torch.no_grad():
		predicted_ids = model.generate(input_features.to("cuda"))[0]
	
	transcription = processor.decode(predicted_ids)
	batch["prediction"] = processor.tokenizer._normalize(transcription)
	
	return batch
	
#Do the evaluation
result = ds.map(map_to_pred)

#Compute the overall WER now.
from evaluate import load

wer = load("wer")
WER=100 * wer.compute(references=result["reference"], predictions=result["prediction"])
print(WER)

Test Result: 7.743795695602924

BibTeX entry and citation info

When publishing results based on these models please refer to:

@inproceedings{mena2024samromur,
  title={Samr{\'o}mur Millj{\'o}n: An ASR Corpus of One Million Verified Read Prompts in Icelandic},
  author={Mena, Carlos Daniel Hernandez and Gunnarsson, {\TH}orsteinn Da{\dh}i and Gu{\dh}nason, J{\'o}n},
  booktitle={Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)},
  pages={14305--14312},
  year={2024}
}

Acknowledgements

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. This model is an unexpected result of all the resources gathered by the Programme.

Special thanks to Björn Ingi Stefánsson for setting up the configuration of the server where this model was trained.