đ Wav2vec 2.0 for Portuguese in 8kHz
This is a fine - tuned model for Portuguese speech recognition, leveraging the power of wav2vec 2.0 technology.
đ Quick Start
This is a fine - tuned model from [facebook/wav2vec2 - base - 10k - voxpopuli](https://huggingface.co/facebook/wav2vec2 - base - 10k - voxpopuli).
đ Documentation
Datasets
The following datasets were used to fine - tune the model:
- CETUC: It contains approximately 145 hours of Brazilian Portuguese speech. The speech is distributed among 50 male and 50 female speakers. Each speaker pronounces approximately 1,000 phonetically balanced sentences selected from the CETEN - Folha corpus.
- Common Voice 7.0: A project proposed by the Mozilla Foundation. Its goal is to create a widely open dataset in different languages. In this project, volunteers donate and validate speech through the official site.
- Lapsbm: "Falabrasil - UFPA" is a dataset used by the Fala Brasil group to benchmark ASR systems in Brazilian Portuguese. It contains 35 speakers (10 females), each pronouncing 20 unique sentences, resulting in a total of 700 utterances in Brazilian Portuguese. The audios were recorded at 22.05 kHz without environment control.
- Multilingual Librispeech (MLS): A massive dataset available in many languages. It is based on audiobook recordings in the public domain like LibriVox. The dataset has a total of 6k hours of transcribed data in many languages. The Portuguese set used in this work (mostly the Brazilian variant) has approximately 284 hours of speech, obtained from 55 audiobooks read by 62 speakers.
- Multilingual TEDx: A collection of audio recordings from TEDx talks in 8 source languages. The Portuguese set (mostly the Brazilian Portuguese variant) contains 164 hours of transcribed speech.
- Sidney (SID): It contains 5,777 utterances recorded by 72 speakers (20 women) aged from 17 to 59 years old. Information such as place of birth, age, gender, education, and occupation is also available.
- VoxForge: A project aiming to build open datasets for acoustic models. The corpus contains approximately 100 speakers and 4,130 utterances of Brazilian Portuguese, with sample rates ranging from 16kHz to 44.1kHz.
- VoxPopuli
Metrics
Tags
- audio
- speech
- wav2vec2
- pt
- portuguese - speech - corpus
- automatic - speech - recognition
- speech
- PyTorch
đ License
This project is licensed under the apache - 2.0 license.
| Property |
Details |
| Model Type |
Fine - tuned wav2vec 2.0 model for Portuguese in 8kHz |
| Training Data |
CETUC, Common Voice 7.0, Lapsbm, Multilingual Librispeech (MLS), Multilingual TEDx, Sidney (SID), VoxForge, VoxPopuli |
| Metrics |
wer |
| Tags |
audio, speech, wav2vec2, pt, portuguese - speech - corpus, automatic - speech - recognition, speech, PyTorch |
| License |
apache - 2.0 |
%20--%3e%3cdefs%3e%3cstyle%3e%20.st0%20{%20fill:%20%23061b40;%20}%20.st1%20{%20fill:%20%23306af1;%20}%20.st2%20{%20fill:%20%235ce5cf;%20}%20%3c/style%3e%3c/defs%3e%3cg%3e%3cpath%20class='st0'%20d='M55,10.5h9v3h-9v9h12V7.5h-12v3ZM64,19.5h-6v-3h6v3Z'/%3e%3cpolygon%20class='st0'%20points='69%2016.5%2078%2016.5%2078%2019.5%2069%2019.5%2069%2022.5%2081%2022.5%2081%2013.5%2072%2013.5%2072%2010.5%2081%2010.5%2081%207.5%2069%207.5%2069%2016.5'/%3e%3cpolygon%20class='st0'%20points='95%2010.5%2095%207.5%2083%207.5%2083%2022.5%2095%2022.5%2095%2019.5%2086%2019.5%2086%2016.5%2095%2016.5%2095%2013.5%2086%2013.5%2086%2010.5%2095%2010.5'/%3e%3cpath%20class='st0'%20d='M40,1.5v21h11.6l1.4-1.4v-7.6h0c0,0-1.4-1.5-1.4-1.5l1.4-1.4V2.9l-1.4-1.4h-11.6ZM50,19.5h-7v-6h7v6ZM50,10.5h-7v-6h7v6Z'/%3e%3c/g%3e%3cpath%20class='st1'%20d='M23.1,24L14.7,4.8l-4.9,11.2h3.8l-1.8,4H3.3L12.1,0H2C.9,0,0,.9,0,2v20c0,1.1.9,2,2,2h21.1Z'/%3e%3cpath%20class='st2'%20d='M34,0h-16.8l10.6,24h6.2c1.1,0,2-.9,2-2V2C36,.9,35.1,0,34,0ZM32.5,20h-4V4h4v16Z'/%3e%3c/svg%3e)
