SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2207.00952 · 2022

M-Adapter: Modality Adaptation for End-to-End Speech-to-Text Translation

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

End-to-end speech-to-text translation models are often initialized with pre-trained speech encoder and pre-trained text decoder. This leads to a significant training gap between pre-training and fine-tuning, largely due to the modality differences between speech outputs from the encoder and text inputs to the decoder. In this work, we aim to bridge the modality gap between speech and text to improve translation quality. We propose M-Adapter, a novel Transformer-based module, to adapt speech representations to text. While shrinking the speech sequence, M-Adapter produces features desired for speech-to-text translation via modelling global and local dependencies of a speech sequence. Our experimental results show that our model outperforms a strong baseline by up to 1 BLEU score on the Must-C En$\rightarrow$DE dataset.\footnote{Our code is available at https://github.com/mingzi151/w2v2-st.}

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2207.00952")
get_code_for_paper("2207.00952")
have("2207.00952")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2207.00952.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2207.00952.svg)](https://syntology.ai/paper/2207.00952)

Connect an agent — have() is free.