Search papers, labs, and topics across Lattice.
3
0
3
Achieving state-of-the-art kanji reading accuracy, Sarashina2.2-TTS sets a new standard for Japanese speech synthesis while ensuring cross-lingual robustness.
SLMs show improved dialectal understanding when trained on dialectal data, revealing a critical pathway for enhancing performance in diverse linguistic contexts.
Translation-enhanced pre-training not only aligns speech encoders with LLMs but also boosts performance across multiple speech tasks.
