Hugging Face's logo

mlboydaisuke
/
EmbeddingGemma-2-Transcribe-LiteRT

Automatic Speech Recognition
LiteRT
LiteRT
English
on-device
android
speech-recognition
audio
embeddinggemma
Model card Files
xet
 Community
1
EmbeddingGemma-2-Transcribe-LiteRT

Commit History

2 days ago

  1. Card: the Galaxy S26 NPU option with the reader (encoder on the NPU in 10 s windows, reader on GPU fp32 on its own thread): the 54-minute recording in 71.1 s, 45.8x real time, WER 4.328 %

    mlboydaisuke
    Verified signature
    84f653a

7 days ago

  1. Reader v2 (6 layers, 19.6 M, LibriSpeech 960 h): the two reader graphs and FILES.sha256; card with the v2 numbers on the Pixel 8a, the Galaxy S26 rows, and the reader repository's v2 revision

    mlboydaisuke
    Verified signature
    af9fe50

  2. Card: Conversion section (sources and revisions, tool versions, the two commands, the byte-identical rebuild of 2026-10-15)

    mlboydaisuke
    Verified signature
    80106e1

8 days ago

  1. EmbeddingGemma 2 Transcribe for LiteRT: fp16-safe encoder (fp16 weights) for 30 s and 10 s windows, reader v1 graphs, vocab, card

    mlboydaisuke
    Verified signature
    54d0e90

  2. initial commit

    mlboydaisuke
    Verified signature
    175b197