AI Paper Atlas

Research map / Video / Video generation

Speech Driven Tongue Animation

Salvador Medina, Denis Tome, Carsten Stoll, Mark Tiede, Kevin Munhall, Alexander G. Hauptmann, Iain Matthews

CVPR 2022 · Video › Video generation · approach: Talking face generation · 14 citations (OpenAlex)

Check 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
Use Wav2Vec-C features instead of phone features with the same 5-layer bidirectional GRU decoderSpeech-to-tongue animation model · Tongue mocap test splitmean sample error (mm): 2.341 -> 1.757Improved
Use Wav2Vec-C features instead of MFCC features with a 5-layer bidirectional GRU decoderSpeech-to-tongue animation model · Tongue mocap test splitmean sample error (mm): 1.973 -> 1.757Improved
Use Wav2Vec-C features instead of DeepSpeech2 features with a 5-layer bidirectional GRU decoderSpeech-to-tongue animation model · Tongue mocap test splitmean sample error (mm): 2.058 -> 1.757Improved
Use a 5-layer bidirectional GRU instead of a 5-layer unidirectional GRU (ablation)Wav2Vec-C speech-to-tongue animation model · Tongue mocap test splitmean sample error (mm): 1.916 -> 1.757Improved
Use a 5-layer bidirectional GRU instead of a 5-layer bidirectional LSTM (ablation)Wav2Vec-C speech-to-tongue animation model · Tongue mocap test splitmean sample error (mm): 1.927 -> 1.757Improved
Use a 5-layer bidirectional GRU instead of a 2-layer GRU (ablation)Wav2Vec-C speech-to-tongue animation model · Tongue mocap test splitmean sample error (mm): 1.897 -> 1.757Improved
Use a 5-layer bidirectional GRU instead of a Transformer decoder (ablation)Wav2Vec-C speech-to-tongue animation model · Tongue mocap test splitmean sample error (mm): 1.935 -> 1.757Improved

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