language: hr
datasets:
- parlaspeech-hr
tags:
- audio
- automatic-speech-recognition
- parlaspeech
widget:
- example_title: example 1
src: >-
https://huggingface.co/classla/wav2vec2-xls-r-parlaspeech-hr/raw/main/1800.m4a
- example_title: example 2
src: >-
https://huggingface.co/classla/wav2vec2-xls-r-parlaspeech-hr/raw/main/00020578b.flac.wav
- example_title: example 3
src: >-
https://huggingface.co/classla/wav2vec2-xls-r-parlaspeech-hr/raw/main/00020570a.flac.wav
wav2vec2-large-slavic-parlaspeech-hr-lm
This model for Croatian ASR is based on the facebook/wav2vec2-large-slavic-voxpopuli-v2 model and was fine-tuned with 300 hours of recordings and transcripts from the ASR Croatian parliament dataset ParlaSpeech-HR v1.0 and enhanced with a 5-gram language model based on the ParlaMint dataset.
If you use this model, please cite the following paper:
Nikola Ljubešić, Danijel Koržinek, Peter Rupnik, Ivo-Pavao Jazbec. ParlaSpeech-HR -- a freely available ASR dataset for Croatian bootstrapped from the ParlaMint corpus. http://www.lrec-conf.org/proceedings/lrec2022/workshops/ParlaCLARINIII/pdf/2022.parlaclariniii-1.16.pdf
Metrics
Evaluation is performed on the dev and test portions of the ParlaSpeech-HR v1.0 dataset.
| split | CER | WER |
|---|---|---|
| dev | 0.0253 | 0.0556 |
| test | 0.0188 | 0.0430 |
Usage in transformers
Tested with transformers==4.18.0, torch==1.11.0, and SoundFile==0.10.3.post1.
from transformers import Wav2Vec2ProcessorWithLM, Wav2Vec2ForCTC
import soundfile as sf
import torch
import os
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
# load model and tokenizer
processor = Wav2Vec2ProcessorWithLM.from_pretrained(
"classla/wav2vec2-large-slavic-parlaspeech-hr-lm")
model = Wav2Vec2ForCTC.from_pretrained("classla/wav2vec2-large-slavic-parlaspeech-hr-lm")
# download the example wav files:
os.system("wget https://huggingface.co/classla/wav2vec2-large-slavic-parlaspeech-hr-lm/raw/main/00020570a.flac.wav")
# read the wav file
speech, sample_rate = sf.read("00020570a.flac.wav")
input_values = processor(speech, sampling_rate=sample_rate, return_tensors="pt").input_values.cuda()
inputs = processor(speech, sampling_rate=sample_rate, return_tensors="pt")
with torch.no_grad():
logits = model(**inputs).logits
transcription = processor.batch_decode(logits.numpy()).text[0]
# remove the raw wav file
os.system("rm 00020570a.flac.wav")
transcription # 'velik broj poslovnih subjekata poslao je sa minusom velik dio'
Training hyperparameters
In fine-tuning, the following arguments were used:
| arg | value |
|---|---|
per_device_train_batch_size |
16 |
gradient_accumulation_steps |
4 |
num_train_epochs |
8 |
learning_rate |
3e-4 |
warmup_steps |
500 |