Hugging Face's logo

lgris
/
bp500-xlsr

Automatic Speech Recognition
Transformers
PyTorch
Portuguese
wav2vec2
audio
speech
portuguese-speech-corpus
PyTorch
hf-asr-leaderboard
Eval Results
Inference Endpoints
Model card Files Community
bp500-xlsr / README.md
lgris's picture
lgris
Create README.md
3309b4b
preview code
|
raw
history blame
12.3 kB
metadata
language: pt
datasets:
  - common_voice
  - mls
  - cetuc
  - lapsbm
  - voxforge
  - tedx
  - sid
metrics:
  - wer
tags:
  - audio
  - speech
  - wav2vec2
  - pt
  - portuguese-speech-corpus
  - automatic-speech-recognition
  - speech
  - PyTorch
license: apache-2.0

bp500-xlsr: Wav2vec 2.0 with Brazilian Portuguese (BP) Dataset

This is a the demonstration of a fine-tuned Wav2vec model for Brazilian Portuguese using the following datasets:

  • CETUC: contains approximately 145 hours of Brazilian Portuguese speech distributed among 50 male and 50 female speakers, each pronouncing approximately 1,000 phonetically balanced sentences selected from the CETEN-Folha corpus;
  • Common Voice 7.0: is a project proposed by Mozilla Foundation with the goal to create a wide open dataset in different languages. In this project, volunteers donate and validate speech using the oficial site;
  • Lapsbm: "Falabrasil - UFPA" is a dataset used by the Fala Brasil group to benchmark ASR systems in Brazilian Portuguese. Contains 35 speakers (10 females), each one pronouncing 20 unique sentences, totalling 700 utterances in Brazilian Portuguese. The audios were recorded in 22.05 kHz without environment control;
  • Multilingual Librispeech (MLS): a massive dataset available in many languages. The MLS is based on audiobook recordings in public domain like LibriVox. The dataset contains a total of 6k hours of transcribed data in many languages. The set in Portuguese used in this work (mostly Brazilian variant) has approximately 284 hours of speech, obtained from 55 audiobooks read by 62 speakers;
  • VoxForge: is a project with the goal to build open datasets for acoustic models. The corpus contains approximately 100 speakers and 4,130 utterances of Brazilian Portuguese, with sample rates varying from 16kHz to 44.1kHz.

These datasets were combined to build a larger Brazilian Portuguese dataset. All data was used for training except Common Voice dev/test sets, that were used for validation/test respectively. We also made test sets for all the gathered datasets.

Dataset Train Valid Test
CETUC 93.9h -- 5.4h
Common Voice 37.6h 8.9h 9.5h
LaPS BM 0.8h -- 0.1h
MLS 161.0h -- 3.7h
Multilingual TEDx (Portuguese) 144.2h -- 1.8h
SID 5.0h -- 1.0h
VoxForge 2.8h -- 0.1h
Total 437.2h 8.9h 21.6h

The original model was fine-tuned using fairseq. This notebook uses a converted version of the original one. The link to the original fairseq model is available here.

Summary

CETUC CV LaPS MLS SID TEDx VF AVG
bp_500 (demonstration below) 0.051 0.136 0.032 0.118 0.095 0.248 0.082 0.108
bp_500 + 4-gram (demonstration below) 0.032 0.097 0.022 0.114 0.125 0.246 0.065 0.100

Transcription examples

Text Transcription
não há um departamento de mediadores independente das federações e das agremiações não há um dearamento de mediadores independente das federações e das agrebiações
mas que bodega masque bodega
a cortina abriu o show começou a cortina abriu o chô começou
por sorte havia uma passadeira busote avinhoa passadeiro
estou maravilhada está tudo pronto stou estou maravilhada está tudo pronto

Demonstration 

MODEL_NAME = "lgris/bp500-xlsr"

Imports and dependencies 

%%capture
!pip install torch==1.8.2+cu111 torchvision==0.9.2+cu111 torchaudio===0.8.2 -f https://download.pytorch.org/whl/lts/1.8/torch_lts.html
!pip install datasets
!pip install jiwer
!pip install transformers
!pip install soundfile
!pip install pyctcdecode
!pip install https://github.com/kpu/kenlm/archive/master.zip

import jiwer
import torchaudio
from datasets import load_dataset, load_metric
from transformers import (
    Wav2Vec2ForCTC,
    Wav2Vec2Processor,
)
from pyctcdecode import build_ctcdecoder
import torch
import re
import sys

Helpers 

chars_to_ignore_regex = '[\,\?\.\!\;\:\"]'  # noqa: W605

def map_to_array(batch):
    speech, _ = torchaudio.load(batch["path"])
    batch["speech"] = speech.squeeze(0).numpy() 
    batch["sampling_rate"] = 16_000 
    batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower().replace("’", "'")
    batch["target"] = batch["sentence"]
    return batch

def calc_metrics(truths, hypos):
    wers = []
    mers = []
    wils = []
    for t, h in zip(truths, hypos):
        try:
            wers.append(jiwer.wer(t, h))
            mers.append(jiwer.mer(t, h))
            wils.append(jiwer.wil(t, h))
        except: # Empty string?
            pass
    wer = sum(wers)/len(wers)
    mer = sum(mers)/len(mers)
    wil = sum(wils)/len(wils)
    return wer, mer, wil

def load_data(dataset):
    data_files = {'test': f'{dataset}/test.csv'}
    dataset = load_dataset('csv', data_files=data_files)["test"]
    return dataset.map(map_to_array)

Model 

class STT:

    def __init__(self, 
                 model_name, 
                 device='cuda' if torch.cuda.is_available() else 'cpu', 
                 lm=None):
        self.model_name = model_name
        self.model = Wav2Vec2ForCTC.from_pretrained(model_name).to(device)
        self.processor = Wav2Vec2Processor.from_pretrained(model_name)
        self.vocab_dict = self.processor.tokenizer.get_vocab()
        self.sorted_dict = {
            k.lower(): v for k, v in sorted(self.vocab_dict.items(), 
                                            key=lambda item: item[1])
        }
        self.device = device
        self.lm = lm
        if self.lm:            
            self.lm_decoder = build_ctcdecoder(
                list(self.sorted_dict.keys()),
                self.lm
            )

    def batch_predict(self, batch):
        features = self.processor(batch["speech"], 
                                  sampling_rate=batch["sampling_rate"][0], 
                                  padding=True, 
                                  return_tensors="pt")
        input_values = features.input_values.to(self.device)
        attention_mask = features.attention_mask.to(self.device)
        with torch.no_grad():
            logits = self.model(input_values, attention_mask=attention_mask).logits
        if self.lm:
            logits = logits.cpu().numpy()
            batch["predicted"] = []
            for sample_logits in logits:
                batch["predicted"].append(self.lm_decoder.decode(sample_logits))
        else:
            pred_ids = torch.argmax(logits, dim=-1)
            batch["predicted"] = self.processor.batch_decode(pred_ids)
        return batch

Download datasets 

%%capture
!gdown --id 1HFECzIizf-bmkQRLiQD0QVqcGtOG5upI
!mkdir bp_dataset
!unzip bp_dataset -d bp_dataset/

%cd bp_dataset 

/content/bp_dataset

Tests 

stt = STT(MODEL_NAME)

CETUC 

ds = load_data('cetuc_dataset')
result = ds.map(stt.batch_predict, batched=True, batch_size=8) 
wer, mer, wil = calc_metrics(result["sentence"], result["predicted"])
print("CETUC WER:", wer)

CETUC WER: 0.05159097808687998

Common Voice 

ds = load_data('commonvoice_dataset')
result = ds.map(stt.batch_predict, batched=True, batch_size=8) 
wer, mer, wil = calc_metrics(result["sentence"], result["predicted"])
print("CV WER:", wer)

CV WER: 0.13659981509705973

LaPS 

ds = load_data('lapsbm_dataset')
result = ds.map(stt.batch_predict, batched=True, batch_size=8) 
wer, mer, wil = calc_metrics(result["sentence"], result["predicted"])
print("Laps WER:", wer)

Laps WER: 0.03196969696969697

MLS 

ds = load_data('mls_dataset')
result = ds.map(stt.batch_predict, batched=True, batch_size=8) 
wer, mer, wil = calc_metrics(result["sentence"], result["predicted"])
print("MLS WER:", wer)

MLS WER: 0.1178481066463896

SID 

ds = load_data('sid_dataset')
result = ds.map(stt.batch_predict, batched=True, batch_size=8) 
wer, mer, wil = calc_metrics(result["sentence"], result["predicted"])
print("Sid WER:", wer)

Sid WER: 0.09544588416964224

TEDx 

ds = load_data('tedx_dataset')
result = ds.map(stt.batch_predict, batched=True, batch_size=8) 
wer, mer, wil = calc_metrics(result["sentence"], result["predicted"])
print("TEDx WER:", wer)

TEDx WER: 0.24868046340420813

VoxForge 

ds = load_data('voxforge_dataset')
result = ds.map(stt.batch_predict, batched=True, batch_size=8) 
wer, mer, wil = calc_metrics(result["sentence"], result["predicted"])
print("VoxForge WER:", wer)

VoxForge WER: 0.08246076839826841

Tests with LM 

!rm -rf ~/.cache
!gdown --id 1GJIKseP5ZkTbllQVgOL98R4yYAcIySFP  # trained with wikipedia
stt = STT(MODEL_NAME, lm='pt-BR-wiki.word.4-gram.arpa')
# !gdown --id 1dLFldy7eguPtyJj5OAlI4Emnx0BpFywg  # trained with bp
# stt = STT(MODEL_NAME, lm='pt-BR.word.4-gram.arpa')

Cetuc 

ds = load_data('cetuc_dataset')
result = ds.map(stt.batch_predict, batched=True, batch_size=8) 
wer, mer, wil = calc_metrics(result["sentence"], result["predicted"])
print("CETUC WER:", wer)

CETUC WER: 0.03222801788375573

Common Voice 

ds = load_data('commonvoice_dataset')
result = ds.map(stt.batch_predict, batched=True, batch_size=8) 
wer, mer, wil = calc_metrics(result["sentence"], result["predicted"])
print("CV WER:", wer)

CV WER: 0.09713866021093655

LaPS 

ds = load_data('lapsbm_dataset')
result = ds.map(stt.batch_predict, batched=True, batch_size=8) 
wer, mer, wil = calc_metrics(result["sentence"], result["predicted"])
print("Laps WER:", wer)

Laps WER: 0.022310606060606065

MLS 

ds = load_data('mls_dataset')
result = ds.map(stt.batch_predict, batched=True, batch_size=8) 
wer, mer, wil = calc_metrics(result["sentence"], result["predicted"])
print("MLS WER:", wer)

MLS WER: 0.11408590958696524

SID 

ds = load_data('sid_dataset')
result = ds.map(stt.batch_predict, batched=True, batch_size=8) 
wer, mer, wil = calc_metrics(result["sentence"], result["predicted"])
print("Sid WER:", wer)

Sid WER: 0.12502797252979136

TEDx 

ds = load_data('tedx_dataset')
result = ds.map(stt.batch_predict, batched=True, batch_size=8) 
wer, mer, wil = calc_metrics(result["sentence"], result["predicted"])
print("TEDx WER:", wer)

TEDx WER: 0.24603179403904793

VoxForge 

ds = load_data('voxforge_dataset')
result = ds.map(stt.batch_predict, batched=True, batch_size=8) 
wer, mer, wil = calc_metrics(result["sentence"], result["predicted"])
print("VoxForge WER:", wer)

VoxForge WER: 0.06542207792207791