metadata

language: hr
datasets:
  - parlaspeech-hr
tags:
  - audio
  - automatic-speech-recognition
  - parlaspeech
widget:
  - example_title: example 1
    src: >-
      https://huggingface.co./5roop/wav2vec2-xls-r-parlaspeech-hr-lm/raw/main/1800.m4a
  - example_title: example 2
    src: >-
      https://huggingface.co./5roop/wav2vec2-xls-r-parlaspeech-hr-lm/raw/main/00020578b.flac.wav
  - example_title: example 3
    src: >-
      https://huggingface.co./5roop/wav2vec2-xls-r-parlaspeech-hr-lm/raw/main/00020570a.flac.wav

wav2vec2-xls-r-parlaspeech-hr-lm

This model for Croatian ASR is based on the facebook/wav2vec2-xls-r-300m model and was fine-tuned with 300 hours of recordings and transcripts from the ASR Croatian parliament dataset ParlaSpeech-HR v1.0.

The efforts resulting in this model were coordinated by Nikola Ljubešić, the rough manual data alignment was performed by Ivo-Pavao Jazbec, the method for fine automatic data alignment from Plüss et al. was applied by Vuk Batanović and Lenka Bajčetić, the transcripts were normalised by Danijel Korzinek, while the final modelling was performed by Peter Rupnik.

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. Submitted to ParlaCLARIN@LREC.

Metrics

split	CER	WER
dev	0.0448	0.1129
test	0.0363	0.0985

Usage in `transformers`

So far untested approach that worked before:

from transformers import Wav2Vec2Processor, 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 = Wav2Vec2Processor.from_pretrained(
    "classla/wav2vec2-xls-r-parlaspeech-hr")
model = Wav2Vec2ForCTC.from_pretrained("classla/wav2vec2-xls-r-parlaspeech-hr")


# download the example wav files:
os.system("wget https://huggingface.co./classla/wav2vec2-xls-r-parlaspeech-hr/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.to(device)

# remove the raw wav file
os.system("rm 00020570a.flac.wav")

# retrieve logits
logits = model.to(device)(input_values).logits

# take argmax and decode
predicted_ids = torch.argmax(logits, dim=-1)
transcription = processor.decode(predicted_ids[0]).lower()

# transcription: 'veliki broj poslovnih subjekata posluje 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

wav2vec2-xls-r-parlaspeech-hr-lm

Metrics

Usage in transformers

Training hyperparameters

Usage in `transformers`