Search papers, labs, and topics across Lattice.
Affiliation:
3
0
4
1
High-performing ASR models can produce accurate transcripts even from contradictory audio, revealing a troubling disconnect between benchmark scores and real-world performance.
Voice AI systems may excel in one area while failing in others, revealing a critical need for multidimensional evaluation beyond traditional benchmarks.
Achieve state-of-the-art TTS and SLM performance while slashing inference costs and eliminating content hallucinations by synchronizing text and acoustic tokens.
