firdavsus/Gemma-4-E4B_hidden_to_audio_tokens-2.0
Gemma-4 S2S Alignment Dataset (Multilingual) – Version 2.0 (260K) This dataset is specifically engineered to train a lightweight, low-latency Hidden-to-Speech (H2S) alignment model. By capturing the raw, abstract semantic representations from the 33rd hidden states of a Text LLM (Gemma-4 8B) and mapping them directly onto quantized discrete audio streams, this dataset bypasses traditional text generation bottlenecks to establish native Speech-to-Speech (S2S) processing… See the full description on the dataset page: https://huggingface.co/datasets/firdavsus/Gemma-4-E4B_hidden_to_audio_tokens-2.0.
042
Nothing at this path on main. The folder may be empty, or the revision may not exist.
