AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Multimodal and Multilingual Embeddings for Large-Scale Speech Mining

Paul-Ambroise Duquenne, Hongyu Gong, Holger Schwenk

NeurIPS 2021 · Large language models › Language, safety and interpretability · approach: Sign and cross-lingual translation · 21 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add mined speech-translation data to trainingLNA bilingual S2T · CoVoST2 De-EnBLEU: 24.4 -> 26.4Improved
add mined speech-translation data to trainingLNA bilingual S2T · CoVoST2 Es-EnBLEU: 29.9 -> 31.6Improved
add mined speech-translation data to trainingLNA bilingual S2T · CoVoST2 Fr-EnBLEU: 30.7 -> 32.0Improved
add mined data to MuST-C trainingS2T Transformer · MuST-C en-esBLEU: 27.2 -> 28.7Improved
add mined data to MuST-C trainingS2T Transformer · MuST-C en-frBLEU: 32.9 -> 34.4Improved
add mined data to MuST-C trainingS2T Transformer · MuST-C en-ruBLEU: 15.3 -> 16.1Improved
add mined data to MuST-C trainingS2T Transformer · MuST-C en-viBLEU: 21.4 -> 21.8Improved
use separate per-language speech encoder instead of one multilingual encoder (ablation)speech encoder · CoVoST2 ru audio vs en translationserror rate (%): 40.54 -> 6.86Improved
use separate per-language speech encoder instead of one multilingual encoder (ablation)speech encoder · CoVoST2 de/fr/es audio vs en translationserror rate (%): fr 2.05 -> 1.86; de 3.36 -> 3.73; es 1.66 -> 1.78Mixed
train with both transcription and translation teachers instead of one (ablation)multilingual speech encoder · CoVoST2 audio vs en translations (de, fr, es)error rate (%): de 3.58/4.06 -> 3.36Improved
train with both teachers instead of one (ablation)multilingual speech encoder · CoVoST2 ru audio vs transcriptions/translationserror rate (%): 25.63 -> 36.19 (transcriptions)Hurt
increase speech encoder training data (ablation)French speech encoder · CoVoST2 fr similarity searcherror rateImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy