GOAT-SLM is a spoken language model with a shared text/speech dual-head architecture that detects paralinguistic cues and generates adaptive spoken responses, outperforming open baselines on the authors' TELEVAL benchmark.
Speechgpt 2.0-preview,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
GOAT-SLM: A Spoken Language Model with Paralinguistic and Speaker Characteristic Awareness
GOAT-SLM is a spoken language model with a shared text/speech dual-head architecture that detects paralinguistic cues and generates adaptive spoken responses, outperforming open baselines on the authors' TELEVAL benchmark.