Tag
A developer documents building an open-source Hinglish text-to-speech system that outperforms existing models by fixing upstream inference bugs and adding a lightweight preprocessing wrapper, achieving high quality without training or GPU resources.
MiniMind-O has released an end-to-end omnimodal model with only 0.1B parameters, supporting text, speech, and image inputs as well as streaming speech output. The project opensources the code, weights, training data, and technical report, emphasizing that both training and inference can be performed quickly on standard GPUs.
This paper introduces MultiLinguahah, an unsupervised multilingual method for acoustic laughter segmentation using Isolation Forests on BYOL-A encoder representations. The authors demonstrate that their approach outperforms state-of-the-art supervised methods in non-English settings by treating laughter detection as an anomaly detection task.
PersonaKit is an open-source web platform designed for rapid prototyping and user testing of diverse personas in full-duplex dialogue systems. It allows researchers to configure persona-specific turn-taking behaviors via JSON and conduct A/B surveys to evaluate sociolinguistic interactions.
Researchers from KTH Royal Institute of Technology propose a two-stage framework that fine-tunes LLMs on dialogue transcripts and uses contrastive learning to create joint embeddings for aligning backchannel signals with conversational context, demonstrating improved context-backchannel retrieval compared to previous methods.
easyaligner is an open-source forced alignment library with GPU acceleration and flexible text normalization that works with all wav2vec2 models on Hugging Face Hub. It addresses practical workflows like handling partial transcripts, irrelevant speech segments, and long audio without chunking while preserving original text formatting.