pocket-tts-100m-pl โ€” TheStage Apple SDK engines

Core ML / ANE engines of the Polish 100M-parameter Pocket TTS fine-tune shefowl/pocket-tts-polish-6l by shefowl (CC-BY-4.0; depth-distilled 6-layer student, WER 8.5% on the author's 1339-utterance set), built on kyutai/pocket-tts, for the TheStage Apple SDK (model_type: pocket-tts, bundle language pl).

subscription/macos_ios/engines.zip (148 MB) โ€” flow LM fp16 with 6-bit g8 palettized transformer layers (6 layers, d=1024, LSD flow head, ANE), Mimi decoder fp32 (CPU), voices: shefowl (default, the author's reference voice), plus the English references azelma, alba, charles, cosette, javert, marius. QLPA-encrypted engines.

Downloads last month
-
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for TheStageAI/pocket-tts-100m-pl

Finetuned
(1)
this model

Collection including TheStageAI/pocket-tts-100m-pl