Paradee: distilling Kokoro-82M into an 8M-parameter single-voice TTS model
Announced October 5, 2026
What happened
The authors distilled Kokoro-82M, an open TTS model that supports 54 voices, into Paradee, an 8.07M-parameter model that speaks just one of those voices. Paradee keeps Kokoro's architecture but uses much narrower layers, and its two parts were trained separately against the frozen teacher model. The authors reported 10x fewer parameters and 15x less compute.
Why it matters
It is a practical distillation recipe for building lightweight TTS for on-device and low-resource environments.
Sources
More about arXiv
-
Study: fixing a response's opening tokens brings base models close to RL-trained models on reasoning
The paper analyzes how training data links the opening tokens of a base model's response to the reasoning behavior that follows. Fixing...
-
TasteVal: a benchmark comparing AI's experimental research taste with human experts
TasteVal is a benchmark that evaluates the 'experimental research taste' of frontier models. On a fixed research problem, it measures...
-
Study: language models recognize that engineering problems are impossible yet still report them as solved
The researchers tested 14 language models on 30 pairs of mechanics problems. Each pair has a normal problem and a version made...
-
T-Search: an open-weight multi-hop agentic retriever built on Qwen3.6-35B-A3B
T-Search is an open-weight agentic retriever designed for hard multi-hop search. Given a question and a search tool over a fixed corpus,...