Papers, modelos e datasets em alta no Hugging Face, além do blog oficial — com leitura editorial em português.
Flexible Spoken Language Model (FlexiSLM) introduces dynamic frame rate capabilities for speech input and output, achieving superior performance over fixed-frame-rate models while…
South Korea targets physical AI lead and commercial humanoid robots by 2028.
O ConvFill propõe um truque de inferência para o dilema mais cruel da voz conversacional — um modelo pequeno responde na hora enquanto um raciocinador mais lento sussurra a resposta certa no meio da conversa.
arXiv:2606.27380v1 Announce Type: new Abstract: Automated coaching for oral presentations sits at the intersection of computer-assisted pronunciation training (CAPT), prosody modeling, and speech synthesis, yet no prior work has systematically surveyed and compared existing systems along these dimensions. This survey reviews and categorizes automated presentation coaching systems, spanning pronunciation tutors, fluency and prosody coaches, multimodal trainers, and conference Q&A practice tools. ...
arXiv:2606.27627v1 Announce Type: new Abstract: Discrete audio representations have become increasingly popular for building multimodal text-audio systems and integrating audio capabilities into Large Language Models (LLMs). However, numerous studies report performance degradation on various downstream tasks due to information loss during discretization. To address this, we propose a novel approach combining temporally compressed discrete tokens with dimensionality-reduced continuous residuals. ...
arXiv:2606.27717v1 Announce Type: new Abstract: Prosodic emphasis varies across languages, emotions, and speaking styles, yet existing emphasis detection models are largely trained and evaluated on monolingual neutral read speech. We introduce MMEE (Multilingual Multi-Emotion Emphasis), a corpus of 10,000 professionally recorded expressive utterances (14.13 hours) across 7 languages and 34 emotion/style categories, with three-level perceptual labels (10 annotations per sample). We benchmark two ...
arXiv:2606.27698v1 Announce Type: new Abstract: Automatic Speech Recognition systems are notoriously both sensitive to adversarial and benign perturbations. While this has been repeatedly demonstrated using reference datasets, detecting such behaviors in deployed systems is incredibly challenging, due to the absence of oracle knowledge of the true transcription. We demonstrate that employing a certification-inspired mechanism can significantly decrease WER, increase recall, and decrease the Spea...
Dataset com 10 mil – 100 mil exemplos — 86 downloads no Hugging Face. Joyo Kanji Yomi Benchmark A kanji-level pronunciation evaluation benchmark for Japanese TTS, covering all 2,136 Joyo kanji and their 4,378 readings wi…
Dataset com 1 mil – 10 mil exemplos — 5.5 mil downloads no Hugging Face. Vāgdhenu — Sanskrit Chant Corpus A single-speaker Sanskrit chant (pārāyaṇa) recording corpus — classical ślokas chanted with tradition-faith…
A equipe OpenMOSS publicou um modelo de síntese de fala de 4 bilhões de parâmetros que produz áudio em qualidade de estúdio e começa a falar antes de terminar de pensar a frase — um passo concreto rumo à voz sintética aberta em tempo real.
Dataset em destaque no Hugging Face — 20.2 mil downloads. ESB Test Sets: Parquet & Sorted This dataset takes the open-asr-leaderboard/datasets-test-only data and sorts each split by audio length.
Modelo de síntese de voz · 4 B de parâmetros — 133.8 mil downloads e 606 curtidas no Hugging Face.