datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
mixture_ami_synthetic_bigmixture_ami_jpn_synthetic_Bigmixture_callhome_jpn_syntheticmixture_ami_syntheticdcase-5class-3source-mixtures-32k
DCASE 5-Class 3-Source Separation 32k
Dataset Description
This dataset is a collection of 10,000 synthetic audio mixtures designed for the task of audio source separation.
Each audio file is a 10-second, 32kHz mixture containing 3 distinct audio sources from a pool of 5 selected classes. The mixtures were generated with a random Signal-to-Noise Ratio (SNR) between 5 and 20 dB.
This dataset is ideal for training and evaluating models that aim to separate a mixed audio… See the full description on the dataset page: https://huggingface.co/datasets/Kiuyha/dcase-5class-3source-mixtures-32k.mixture_callfriend_jpn_syntheticaudio-flamingo-3-hf_sft_mixture_0.5_advwave_100000_500Qwen2-Audio-7B-Instruct_sft_mixture_0.9_advwaveQwen2-Audio-7B-Instruct_sft_mixture_0.5_advwave_150000_0audio-flamingo-3-hf_sft_mixture_0.5_advwave_50000_500Qwen2-Audio-7B-Instruct_sft_mixture_0.5_advwave_200000Qwen2-Audio-7B-Instruct_sft_mixture_0.2_advwaveQwen2-Audio-7B-Instruct_sft_mixture_1_advwaveQwen2-Audio-7B-Instruct_sft_mixture_0_advwave_50000Qwen2-Audio-7B-Instruct_sft_mixture_0.5_advwave_150000Qwen2-Audio-7B-Instruct_mixture_0Qwen2-Audio-7B-Instruct_sft_mixture_0_advwaveQwen2-Audio-7B-Instruct_sft_mixture_0.1_advwave_50000Qwen2-Audio-7B-Instruct_sft_mixture_0.1_advwave_200000Qwen2-Audio-7B-Instruct_sft_mixture_0.9_advwave_200000Qwen2-Audio-7B-Instruct_sft_mixture_1_advwave_200000Qwen2-Audio-7B-Instruct_mixture_1Qwen2-Audio-7B-Instruct_sft_mixture_0Qwen2-Audio-7B-Instruct_sft_mixture_0.5_advwaveQwen2-Audio-7B-Instruct_mixture_0_advwaveQwen2-Audio-7B-Instruct_mixture_0.999_advwaveQwen2-Audio-7B-Instruct_sft_mixture_0.9_advwave_2000Qwen2-Audio-7B-Instruct_sft_mixture_0.9_advwave_200Qwen2-Audio-7B-Instruct_sft_mixture_0.9_advwave_50000Qwen2-Audio-7B-Instruct_mixture_1_advwave