Search papers, labs, and topics across Lattice.
2
0
4
11
A deployable system for streaming Thai ASR that lets a user steer its vocabulary at decode time, without retraining is presented, which lifts keyword recall from 16.6% to 20.7% at no accuracy cost and negligible overhead.
The resulting 82M-parameter model, Wayu-Paxa-TTS-Edge, enables on-device Thai TTS without reference audio and achieves the lowest pause-placement error and intra-word pause rates among the three systems.
