Hugging Face's logo

spybyscript
/
scyllasband_old

Text-to-Speech
LiteRT
ONNX
LiteRT
English
Spanish
Italian
onnxruntime
tts
int8
quantization
speech-synthesis
duration-flow
multilingual
expressive-tts
emotion
cpu-inference
Model card Files
xet
 Community
1
scyllasband_old

Commit History

  1. Restore original 4-language G2P (document-level regression in the v7 G2P; 85 vs 15 audible edits on the 145-line document suite)

    spybyscript
    Verified signature
    b6e1a03

  1. Fully quantized Core AI bundle (macOS only: iPhone GPU compilation runs out of memory on the int8 vector estimator)

    spybyscript
    Verified signature
    975ff57

  2. Default Core AI bundle: int8 frontends + fp32 vector/vocoder (full-int8 OOMs iPhone GPU compilation; moved to coreai-int8)

    spybyscript
    Verified signature
    db1b503

  3. Keep the fp32 Core AI bundle available as coreai-fp32

    spybyscript
    Verified signature
    d05b2ea

  4. Make the int8-quantized Core AI bundle the default (g2p/duration/context/vector int8, vocoder fp32; 384-frame vocoder bucket excluded pending compiler fix)

    spybyscript
    Verified signature
    26c3761

  5. Add Scylla's Band Core AI bundle for iOS and macOS 27 (384-frame vocoder bucket excluded pending re-export)

    spybyscript
    Verified signature
    ac26809

  1. Scylla's Band v1: G2P v7 (7 languages), b64/b128 latent buckets, model_version 1

    spybyscript
    Verified signature
    a8deaa2

  1. Add optional INT8 ONNX bundle

    spybyscript
    Verified signature
    49abc9a

  1. Initial public release

    spybyscript
    c00d50c