Tag
SignTrace is a system that enables reverse lookup in Chinese sign language dictionaries using large language models, achieving 94.0% Hit@1 accuracy for identifying signs from natural movement descriptions.
Google Gemini now supports sign language transcription from video, a notable accessibility advancement for AI models.
This paper proposes a gloss-free representation learning approach for cross-dataset sign spotting, using weakly aligned broadcast transcripts in Turkish Sign Language. It shows that LLM-assisted pseudo-gloss normalization improves temporal localization and downstream translation quality.
Google DeepMind announced SL2T, a system developed with the Deaf community to bring ASL input to phones, with plans to expand to more sign languages and applications.
Google DeepMind has developed an AI model that translates sign language by directly converting hand, body, and face movements into text, addressing long-standing technical challenges.
DeepMind introduces SL2T, a massively multilingual sign-language-to-text translation model, now powering sign-to-text dictation in Gboard and Live Transcribe on Pixel 11 for American Sign Language, with more languages to follow.
Hand Wave is an open-source sign language recognition system for web and iOS, backed by a Python inference service, aiming to enable translation via smart glasses.
Uvilox AI introduces a system that translates American Sign Language into real-time emergency alerts, improving accessibility for deaf individuals in emergencies.
Hand Wave is a smart glasses product that translates sign language into speech, aimed at improving accessibility for deaf and hard-of-hearing users.
This paper evaluates whether Sign Language Recognition models exhibit phonological sensitivity by probing them with minimal pairs of signs, revealing architectural trade-offs and emergent but limited phonological perception.
This paper presents NEST-V1, a proof-of-concept multimodal framework for generating emotion-conditioned Nepali Sign Language avatars from spoken input, achieving 81.1% ASR accuracy and 79.21% emotion recognition accuracy on a dataset of 600 audio samples from 50 speakers.
Signspell is a Python package for real-time American Sign Language alphabet recognition, installable via pip.
This paper introduces the eJSL Dialog dataset for emotion recognition in sign language conversations, addressing the lack of conversational context in existing datasets. Benchmarking shows a domain gap when applying generic multimodal models, highlighting the need for context-aware visual extractors for sign language.
This paper introduces a direct sign-to-sign translation model that bypasses intermediate text by using back-translation to create synthetic parallel sign language data, achieving significant improvements in speed and accuracy over cascade methods for ASL, CSL, and DGS.
A research paper presenting a dataset and XGBoost-based model for sentiment analysis of German Sign Language (DGS) fairy tales using facial and body motion features extracted via MediaPipe, achieving 63.1% balanced accuracy and demonstrating the importance of both facial and body movements for sentiment communication in sign language.
Ona AI is building digital sign language avatars and inclusive datasets to improve accessibility for deaf and hard of hearing communities.
Ona AI is developing digital sign language avatars and inclusive datasets to create assistive technology for learners with hearing impairments.