Skip to content
ModelListedExternal
Download

Kokoro Coreml

On-device neural TTS for the Kensora iOS app (K509, ADR 0085). Public Kokoro-82M weights re-hosted as staged Core ML models, MLX-free.

KindOtherVersionRevision 07e3fffBgrade

Model source

Kind
Other
File size
1.32–203 MB
Version
vRevision 07e3fff
Updated
Oct 20, 2026
Downloads
575
Source
Hugging Face

Source description

Kokoro-82M — Core ML (staged) + misaki OOV speller

On-device neural TTS for the Kensora iOS app (K509, ADR 0085). Public Kokoro-82M weights re-hosted as staged Core ML models, MLX-free.

  • KokoroFrontA.mlpackage, KokoroProsodyB.mlpackage, KokoroDecoder.mlpackage — the staged synthesis pipeline (front/duration, prosody/F0N, iSTFTNet vocoder).
  • BartEncoder.mlpackage, BartDecoder.mlpackage — misaki's out-of-vocabulary grapheme-to-phoneme speller (tiny BART), for pronouncing words outside the lexicon.

Sources

1 source
Hugging FacePrimaryVersion 07e3fff0ecef68a7db053e3899355003587efbf1
huggingface.co/kensora/kokoro-coreml/tree/07e3fff0ecef68a7db053e3899355003587efbf1

Verified Aug 30

Model artifacts

6 artifacts

BartEncoder.mlpackage/Data/com.apple.CoreML/weights/weight.bin

bin · 1.32 MB · SHA-256 04fd1f3bc214…5f7d · Hugging Face

Download

BartPrefill.mlpackage/Data/com.apple.CoreML/weights/weight.bin

bin · 1.55 MB · SHA-256 eff6c45e3ff5…5aa3 · Hugging Face

Download

BartStep.mlpackage/Data/com.apple.CoreML/weights/weight.bin

bin · 1.58 MB · SHA-256 af2bcf732d25…fba4 · Hugging Face

Download

KokoroDecoder.mlpackage/Data/com.apple.CoreML/weights/weight.bin

bin · 203 MB · SHA-256 bbf4ad11da9b…11a5 · Hugging Face

Download

KokoroFrontA.mlpackage/Data/com.apple.CoreML/weights/weight.bin

bin · 74.3 MB · SHA-256 be52a0c0081c…9299 · Hugging Face

Download

KokoroProsodyB.mlpackage/Data/com.apple.CoreML/weights/weight.bin

bin · 32.1 MB · SHA-256 f012f264e044…a124 · Hugging Face

Download

Source excerpts

2 excerpts

--- license: apache-2.0 library_name: coreml tags: [text-to-speech, kokoro, core-ml, kensora] --- # Kokoro-82M — Core ML (staged) + misaki OOV speller On-device neural TTS for the Kensora iOS app (K509, ADR 0085). Public Kokoro-82M weights re-hosted as staged Core ML models, MLX-free. - `KokoroFrontA.mlpackage`, `KokoroProsodyB.mlpackage`, `KokoroDecoder.mlpackage` — the staged synthesis pipeline (front/duration, prosody/F0N, iSTFTNet vocoder). - `BartEncoder.mlpackage`, `BartDecoder.mlpackage` — misaki's out-of-vocabulary grapheme-to-phoneme speller (tiny BART), for pronouncing words outside the lexicon.

Source context: 575 downloads · 0 likes · Pipeline text-to-speech · Library coreml · Repo kensora/kokoro-coreml