Automatic metrics play a crucial role in machine translation. Despite th...
As a subjective metric to evaluate the quality of synthesized speech, Me...
Parameter sharing has proven to be a parameter-efficient approach. Previ...
Deploying NMT models on mobile devices is essential for privacy, low lat...
How can speech-to-text translation (ST) perform as well as machine
trans...
While diffusion models have achieved great success in generating continu...
Multimodal machine translation (MMT) aims to improve translation quality...
Is it possible to leverage large scale raw and raw parallel corpora to b...
Nearest Neighbor Machine Translation (kNNMT) is a simple and effective m...
How to solve the data scarcity problem for end-to-end speech-to-text
tra...
This report describes our VolcTrans system for the WMT22 shared task on
...
Speech is the surface form of a finite set of phonetic units, which can ...
Domain adaptation is an important challenge for neural machine translati...
Direct Speech-to-speech translation (S2ST) has drawn more and more atten...
How can we learn unified representations for spoken utterances and their...
This paper introduces GigaST, a large-scale pseudo speech translation (S...
How to learn a better speech representation for end-to-end speech-to-tex...
The punctuation restoration task aims to correctly punctuate the output
...
Transformer-based neural models are used in many AI applications. Traini...
This paper describes the Volctrans' submission to the WMT21 news transla...
How to effectively adapt neural machine translation (NMT) models accordi...
This paper presents a unified end-to-end frame-work for both streaming a...
Can pre-trained BERT for one language and GPT for another be glued toget...
This paper presents Self-correcting Encoding (Secoco), a framework that
...
Multilingual Neural Machine Translation (MNMT) has aroused widespread
in...
Existing multilingual machine translation approaches mainly focus on
Eng...
Multilingual neural machine translation aims at learning a single transl...
This paper describes the systems submitted to IWSLT 2021 by the Volctran...
Having numerous potential applications and great impact, end-to-end spee...
End-to-end speech translation models have become a new trend in the rese...
Developing a unified multilingual translation model is a key topic in ma...
Automatic translation of dialogue texts is a much needed demand in many ...
Fine-tuning is a major approach for domain adaptation in Neural Machine
...
NeurST is an open-source toolkit for neural speech translation developed...
Despite the recent success on image classification, self-training has on...
Pre-trained contextual representations like BERT have achieved great suc...
This paper describes our VolcTrans system on WMT20 shared news translati...
In this paper, we describe our submissions to the WMT20 shared task on
p...
Transformer, BERT and their variants have achieved great success in natu...
Discourse context has been proven useful when translating documents. It ...
We investigate the following question for machine translation (MT): can ...
End-to-end speech-to-text translation (ST), which directly translates th...
An end-to-end speech-to-text translation (ST) takes audio in a source
la...
There are thousands of languages on earth, but visual perception is shar...
Non-autoregressive neural machine translation achieves remarkable infere...
Auto-regressive sequence generative models trained by Maximum Likelihood...
This paper proposes the building of Xiaomingbot, an intelligent, multili...
Neural machine translation (NMT) is ineffective for zero-resource langua...
Non-autoregressive models are promising on various text generation tasks...
GPT-2 and BERT demonstrate the effectiveness of using pre-trained langua...