9 results

audio-encoder

inall models datasets apps agents people projects

Models

All models matching “audio-encoder”

Aratako /gemma-4-E4B-it-audio-encoderfeature-extractiontransformers16 likes468 downloads6mo agoHugging Face Atotti /qwen2-audio-encoderfeature-extractiontransformers2 likes173 downloads1y agoHugging Face rnagabh /gemma4-audio-encoderfeature-extractiontransformersen5 likes94 downloads6mo agoHugging Face gwkrsrch2 /qwen2-audio-encoder-from-qwen2-audio-7b-instructfeature-extractiontransformers1 likes76 downloads1y agoHugging Face mesolitica /gemma-3n-e4b-it-audio-encoder4 likes35 downloads1y agoHugging Face FlexiSLM /Qwen2_5-Omni-Audio_Encoderfeature-extractiontransformers0 likes33 downloads2mo agoHugging Face malaysia-ai /gemma3n-audio-encoder-whisper-decoderfeature-extractiontransformers0 likes26 downloads1y agoHugging Face malaysia-ai /gemma3n-audio-encoder-VQ-32k-whisper-decoderfeature-extractiontransformers0 likes25 downloads1y agoHugging Face

Datasets

All datasets matching “audio-encoder”

avaeziaiteam /Audio-Encoder-Training-Datagated Audio-Encoder-Training-Data Stage-1 data for adapting the audio encoder of Gemma 4 E2B to Persian ASR (audio -> Soniox transcript, no draft in the prompt). Three configs, one schema. Sensitive: the call-center part contains real customer calls (names, phone numbers, order details). Labels are machine-generated (Soniox stt-async-v5), not human transcripts. config rows hours train h val h movies 65,917 201.4 197.2 4.2 youtube 41,149 199.7 195.7 4.0 callcenter 64,895… See the full description on the dataset page: https://huggingface.co/datasets/avaeziaiteam/Audio-Encoder-Training-Data.audio100K<n<1M0 likes15 downloads8d agoHugging Face