asa-temporal-sft-timeaudio-best

Best temporal-localization model from the bachelor thesis Automatic Speech Assessment (Schmidt-Svejstrup & Lindahl, DTU 2026). Timestamp-last target with the TimeAudio representation; mean IoU 0.883 across FOR/LIVE/P501. Fine-tuned from Qwen2-Audio-7B. Code: https://github.com/DavidLindahl/automatic-speech-assessment

Downloads last month
40
Safetensors
Model size
8B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support