Towards disentangling the contributions of articulation and acoustics in multimodal phoneme recognition

Open in new window