metadata
language: ru
thumbnail: null
tags:
- automatic-speech-recognition
- CTC
- Attention
- pytorch
- speechbrain
license: apache-2.0
datasets:
- buriy-audiobooks-2-val
metrics:
- wer
- cer
| Release | Test WER | GPUs |
|---|---|---|
| 22-05-11 | - | 1xK80 24GB |
Pipeline description
(by SpeechBrain text)
This ASR system is composed with 3 different but linked blocks:
- Tokenizer (unigram) that transforms words into subword units and trained with the train transcriptions of LibriSpeech.
- Neural language model (RNNLM) trained on the full (380K) words dataset.
- Acoustic model (CRDNN + CTC/Attention). The CRDNN architecture is made of N blocks of convolutional neural networks with normalisation and pooling on the frequency domain. Then, a bidirectional LSTM is connected to a final DNN to obtain the final acoustic representation that is given to the CTC and attention decoders.
The system is trained with recordings sampled at 16kHz (single channel). The code will automatically normalize your audio (i.e., resampling + mono channel selection) when calling transcribe_file if needed.
Install SpeechBrain
First of all, please install SpeechBrain with the following command:
pip install speechbrain
Please notice that SpeechBrain encourage you to read tutorials and learn more about SpeechBrain.
Transcribing your own audio files (in Russian)
from speechbrain.pretrained import EncoderDecoderASR
asr_model = EncoderDecoderASR.from_hparams(source="AndyGo/speech-brain-asr-crdnn-rnnlm-buriy-audiobooks-2-val", savedir="pretrained_models/speech-brain-asr-crdnn-rnnlm-buriy-audiobooks-2-val")
asr_model.transcribe_file('speech-brain-asr-crdnn-rnnlm-buriy-audiobooks-2-val/example.wav')
Inference on GPU
To perform inference on the GPU, add run_opts={"device":"cuda"} when calling the from_hparams method.