Add SDP-pruned, WebGPU-compatible MeloTTS model (v2) — removes GatherND-based stochastic duration predictor; deterministic durations, bit-identical audio on CPU

docs(readme): reflect extension GPU Lite naming (mobile still Plus, same engine)

docs(extension): rename Plus → GPU Lite tier (matches extension UI; mobile still Plus)

docs(readme): cross-platform Melo Plus parity (extension fp32 + mobile fp16)

docs(extension): document Plus tier (neural-melo-en) + engine tier table

melo extension: upstream LICENSE (MIT)

melo extension: tokens (phoneme → id map)

melo extension: lexicon (enriched, incl. em-dash silence rule)

melo extension: upload fp32 model

docs(mobile-android): Melo v2 sections + retention table for legacy zips

docs(readme): add MeloTTS to engine families + 10-zip layout

melo v2: re-roll with em-dash silence rule in lexicon

Upload threadcast-melo-en-v2.zip

Upload threadcast-melo-en-v2.zip

Upload threadcast-melo-en-v2.zip

Upload threadcast-melo-en-v1.zip

Upload README.md

Upload README.md

Upload README.md

Upload threadcast-kitten-nano-en-v1.zip

Upload 27 files

Upload README.md

Upload README.md

Upload README.md

Upload README.md

Upload README.md

Upload 7 files

Upload README.md with huggingface_hub

Upload assets/logo.png with huggingface_hub

Delete sherpa-onnx-wasm-main-tts.wasm

Delete sherpa-onnx-wasm-main-tts.data

Delete piper_phonemize.data

Delete en_US-lessac-medium.onnx.json

Delete en_US-lessac-medium.onnx

Upload 26 files

Upload 2 files

Upload 3 files

initial commit