arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2107.05604cs.CLcs.LGeess.AS
收藏

离散单元的直接语音到语音翻译

Direct speech-to-speech translation with discrete units

Meta AI;Johns Hopkins University(约翰斯·霍普金斯大学)

Ann Lee, Peng-Jen Chen, Changhan Wang, Jiatao Gu, Sravya Popuri, Xutai Ma, Adam Polyak, Yossi Adi, Qing He, Yun Tang, Juan Pino, Wei-Ning Hsu

摘要:

我们提出了一个直接的语音到语音翻译(S2ST)模型,从一种语言的语音翻译到另一种语言的语音,而不依赖于中间文本生成。我们通过首先在目标语音上应用自监督离散语音编码器,然后训练序列到序列语音到单元翻译(S2UT)模型来预测目标语音的离散表示来解决这个问题。当目标文本成绩单可用时,我们设计了一个联合语音和文本训练框架,使模型能够在同一推理过程中同时生成双模态输出(语音和文本)。Fisher西班牙语-英语数据集上的实验表明,与预测谱图特征的基线直接S2ST模型相比,所提出的框架产生6.7 BLEU的改进。当在没有任何文本记录的情况下进行训练时,我们的模型性能与预测频谱图并使用文本监督进行训练的模型相当,这表明我们的系统在非书面语言之间进行翻译的潜力。音频样本可在www.example.com上获得。

英文摘要:

We present a direct speech-to-speech translation (S2ST) model that translates speech from one language to speech in another language without relying on intermediate text generation. We tackle the problem by first applying a self-supervised discrete speech encoder on the target speech and then training a sequence-to-sequence speech-to-unit translation (S2UT) model to predict the discrete representations of the target speech. When target text transcripts are available, we design a joint speech and text training framework that enables the model to generate dual modality output (speech and text) simultaneously in the same inference pass. Experiments on the Fisher Spanish-English dataset show that the proposed framework yields improvement of 6.7 BLEU compared with a baseline direct S2ST model that predicts spectrogram features. When trained without any text transcripts, our model performance is comparable to models that predict spectrograms and are trained with text supervision, showing the potential of our system for translation between unwritten languages. Audio samples are available at https://facebookresearch.github.io/speech_translation/direct_s2st_units/index.html .

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑