Unsupervised Cross-lingual Representation Learning for Speech Recognition
This enables competitive multilingual speech recognition, particularly benefiting low-resource languages by reducing the need for labeled data.
The paper tackles cross-lingual speech recognition by pretraining a single model on raw speech waveforms from multiple languages, achieving a 72% relative reduction in phoneme error rate on CommonVoice and a 16% relative improvement in word error rate on BABEL compared to prior methods.
This paper presents XLSR which learns cross-lingual speech representations by pretraining a single model from the raw waveform of speech in multiple languages. We build on wav2vec 2.0 which is trained by solving a contrastive task over masked latent speech representations and jointly learns a quantization of the latents shared across languages. The resulting model is fine-tuned on labeled data and experiments show that cross-lingual pretraining significantly outperforms monolingual pretraining. On the CommonVoice benchmark, XLSR shows a relative phoneme error rate reduction of 72% compared to the best known results. On BABEL, our approach improves word error rate by 16% relative compared to a comparable system. Our approach enables a single multilingual speech recognition model which is competitive to strong individual models. Analysis shows that the latent discrete speech representations are shared across languages with increased sharing for related languages. We hope to catalyze research in low-resource speech understanding by releasing XLSR-53, a large model pretrained in 53 languages.