Wav2Letter: an End-to-End ConvNet-based Speech Recognition System
Abstract
An end-to-end speech recognition model combines a convolutional acoustic network with graph decoding, trains on letters from transcribed speech without phoneme alignment, and uses a simpler alignment-free segmentation criterion comparable to CTC, achieving competitive word error rates on Librispeech with MFCC features and raw waveform.
This paper presents a simple end-to-end model for speech recognition, combining a convolutional network based acoustic model and a graph decoding. It is trained to output letters, with transcribed speech, without the need for force alignment of phonemes. We introduce an automatic segmentation criterion for training from sequence annotation without alignment that is on par with CTC while being simpler. We show competitive results in word error rate on the Librispeech corpus with MFCC features, and promising results from raw waveform.
Get this paper in your agent:
hf papers read 1609.03193 Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash Models citing this paper 1
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper