Japanese SFT/DPO data convert to speech via TTS. And audio caption data generated by Qwen3-Omni. All datasets are available for commercial use.
Ayuto Tsutsumi
Atotti
AI & ML interests
None yet
Recent Activity
liked
a model 3 days ago
StyleDistance/mstyledistance liked
a model 4 days ago
llm-jp/llm-jp-moshi-v1 liked
a model 13 days ago
sbintuitions/diafill-llm-jp-3.1-13b-instruct4 Organizations
models 30
Atotti/xacle-tmu-2026
Audio-Text-to-Text • Updated
• 43
Atotti/LlamaForSpeechLM-ja-Instruct-Full-step10000
9B • Updated
Atotti/LlamaForSpeechLM-ja-Instruct-Full-step20000
9B • Updated
Atotti/LlamaForSpeechLM-ja-DPO-Full-step3500
9B • Updated
Atotti/LlamaForSpeechLM-ja-Instruct-Full-step37000
9B • Updated
Atotti/LlamaForSpeechLM-ja-DPO-Full-step8000
9B • Updated
• 17
Atotti/LlamaForSpeechLM-ja-Transcribe-Full-step200000
9B • Updated
Atotti/LlamaForSpeechLM-ja-DPO-Full-step1000
9B • Updated
Atotti/LlamaForSpeechLM-ja-Instruct-Full-step30000
9B • Updated
• 7
Atotti/LlamaForSpeechLM-ja-Instruct-20260112-223832-step11000
9B • Updated
datasets 9
Atotti/fsd50k-ccby-Qwen3-Omni-captioned
Viewer
• Updated
• 23.5k • 233
Atotti/spoken-multiturn-sft
Viewer
• Updated
• 59.1k • 550
Atotti/spoken-dpo-49k
Viewer
• Updated
• 43.1k • 232
Atotti/fsd50k-cc0-Qwen3-Omni-captioned
Viewer
• Updated
• 19.9k • 248
Atotti/spoken-magpie-ja
Viewer
• Updated
• 132k • 388
Atotti/audiocaps-ja
Viewer
• Updated
• 57.2k • 14
Atotti/clotho-ja
Viewer
• Updated
• 19.7k • 40
Atotti/VTuber-overview-split
Viewer
• Updated
• 2.94k • 29 • 1
Atotti/VTuber-overview
Viewer
• Updated
• 420 • 5