Onsei IOS Models
ONNX model packs exported for the Onsei iOS client from the following upstream models:
Modellquelle
- Art
- Other
- Dateigröße
- 65,1 MB–1,19 GB
- Version
- vRevision 158b33f
- Aktualisiert
- 19. Okt. 2026
- Downloads
- 87
- Quelle
- Hugging Face
Quellenbeschreibung
Onsei iOS ONNX and Core ML Model Packs
ONNX model packs exported for the Onsei iOS client from the following upstream models:
Aratako/Irodori-TTS-v4.1-Small-MF, source revisionccc78f5d480b6e51b69b2d5042a14c4da04fea6eAratako/Irodori-TTS-v4.1-Small, source revision2b28324dc263ed5e6638b3cf3dd94c82ead07b4bAratako/Irodori-TTS-500M-v3, source revision236c1e56591279fc24e3c1bf6609fc06e48dde28Aratako/Irodori-TTS-600M-v3-VoiceDesign, source revisione863a3a93e652e09afeff3e84823a206a0a60314Aratako/Semantic-DACVAE-Japanese-32dim- Tokenizer from
llm-jp/llm-jp-3-150m
These files are converted runtime artifacts, not new model weights. They are not affiliated with or endorsed by the upstream authors.
Packs
v4.1-small-mf/: MeanFlow-distilled v4.1 Small. It preserves text, Voice Design, reference-audio cloning and duration conditioning while using one conditional DiT evaluation per step. The app defaults to four steps and ignores RF-only CFG, sway, rescale and speaker-KV controls.v4.1-small/: unified Japanese TTS, reference cloning and VoiceDesign, with the improved v4.1 duration predictor.text_encoder.onnxcontains the shared ModernBERT backbone and bothtext_stateandcaption_stateoutputs. The app reuses one session for both inputs. No separatecaption_encoder.onnxis needed. Requires the v4.1-compatible iOS client.500m-v3/: standard Japanese TTS and reference-audio voice cloning.600m-v3-vd/: VoiceDesign model with caption conditioning.
Each pack contains the split context_kv.onnx + dit_step.onnx execution path used by the Onsei iOS app, DACVAE encode/decode, speaker and text encoders, duration prediction, configuration, and tokenizer data. manifest.json records file sizes and SHA-256 digests used by the app to verify downloads.
Licenses and attribution
The Irodori-TTS weights, Japanese Semantic-DACVAE adaptations, ModernBERT-ja-310m encoder/tokenizer and conversion code declare MIT. The underlying Meta DACVAE weights/implementation declare Apache License 2.0; Descript DAC uses MIT. The legacy llm-jp tokenizer uses Apache License 2.0. See LICENSES/ and THIRD_PARTY_NOTICES.md. Each component remains subject to its own license; no relicensing is implied.
Responsible use
The upstream model cards request that users:
- Do not clone or impersonate a person's voice without explicit consent.
- Do not create misleading deepfakes or synthetic speech intended to spread misinformation.
- Recognize that text-only output may coincidentally resemble a real voice.
- Take responsibility for compliance with applicable law.
The ONNX export used here does not add a SilentCipher watermark. Applications using these files should clearly disclose synthetic audio where appropriate and must enforce consent for voice cloning.
Citation
@misc{irodori-tts-v3,
author = {Chihiro Arata},
title = {Irodori-TTS: A Flow Matching-based Text-to-Speech Model with Emoji-driven Style Control},
year = {2026},
publisher = {Hugging Face},
howpublished = {https://huggingface.co/Aratako/Irodori-TTS-500M-v3}
}
Native GPU decoder (ios-v3 and later)
The v4 packs contain the verified float32 Core ML decoder. New clients download it together with the ONNX pack and verify each file using SHA-256. The app measures CPU/GPU speed and numerical parity on the device before selecting the faster decoder. Existing immutable release tags remain unchanged. The sidecar includes licenses, attribution and conversion notices.
Quellen
1 Quelle
Quellen
1 QuelleVerifiziert 19. Okt.
Modellartefakte
31 Artefakte
Modellartefakte
31 Artefakte600m-v3-vd/caption_encoder.onnx
onnx · 323 MB · SHA-256 c429f84f5beb…9ced · Hugging Face
Herunterladen600m-v3-vd/speaker_encoder.onnx
onnx · 231 MB · SHA-256 8b3412ec1ba0…a859 · Hugging Face
Herunterladenv4.1-small-mf/dacvae_decode.onnx
onnx · 249 MB · SHA-256 48f55372b800…1e48 · Hugging Face
Herunterladenv4.1-small-mf/dacvae_encode.onnx
onnx · 104 MB · SHA-256 3ccc52d3a5b7…4169 · Hugging Face
Herunterladenv4.1-small-mf/native_decoder/decoder.mlpackage/Data/com.apple.CoreML/weights/weight.bin
bin · 249 MB · SHA-256 738aef9a7e29…c12a · Hugging Face
Herunterladenv4.1-small-mf/speaker_encoder.onnx
onnx · 232 MB · SHA-256 ecfa3a47f0a5…945f · Hugging Face
Herunterladenv4.1-small-mf/text_encoder.onnx
onnx · 1,19 GB · SHA-256 961efbf4b01e…2f39 · Hugging Face
Herunterladenv4.1-small/native_decoder/decoder.mlpackage/Data/com.apple.CoreML/weights/weight.bin
bin · 249 MB · SHA-256 738aef9a7e29…c12a · Hugging Face
Herunterladenv4.1-small/speaker_encoder.onnx
onnx · 232 MB · SHA-256 ecfa3a47f0a5…945f · Hugging Face
HerunterladenQuellenauszüge
3 Auszüge
Quellenauszüge
3 Auszüge--- license: other language: - ja pipeline_tag: text-to-speech tags: - coreml - onnx - ios - on-device - irodori-tts --- # Onsei iOS ONNX and Core ML Model Packs ONNX model packs exported for the Onsei iOS client from the following upstream models: - [`Aratako/Irodori-TTS-v4.1-Small-MF`](https://huggingface.co/Aratako/Irodori-TTS-v4.1-Small-MF), source revision `ccc78f5d480b6e51b69b2d5042a14c4da04fea6e` - [`Aratako/Irodori-TTS-v4.1-Small`](https://huggingface.co/Aratako/Irodori-TTS-v4.1-Small), source revision `2b28324dc263ed5e6638b3cf3dd94c82ead07b4b` - [`Aratako/Irodori-TTS-500M-v3`](https://huggingface.co/Aratako/Irodori-TTS-500M-v3), source revision `236c1e56591279fc24e3c1bf6609fc06e48dde28` - [`Aratako/Irodori-TTS-600M-v3-VoiceDesign`](https://huggingface.co/Aratako/Irodori-TTS-600M-v3-VoiceDesign), source revision `e863a3a93e652e09afeff3e84823a206a0a60314` - [`Aratako/Semantic-DACVAE-Japanese-32dim`](https://huggingface.co/Aratako/Semantic-DACVAE-Japanese-32dim) - Tokenizer from [`llm-jp/llm-jp-3-150m`](https://huggingface.co/llm-jp/llm-jp-3-150m) These files are converted runtime artifacts, not new model weights. They are not affiliated with or endorsed by the upstream authors. ## Packs - `v4.1-small-mf/`: MeanFlow-distilled v4.1 Small. It preserves text, Voice Design, reference-audio cloning and duration conditioning while using one conditional DiT evaluation per step. The app defaults to four steps and ignores RF-only CFG, sway, rescale and speaker-KV controls. - `v4.1-small/`: unified Japanese TTS, reference cloning and VoiceDesign, with the improved v4.1 duration predictor. `text_encoder.onnx` contains the shared ModernBERT backbone and both `text_state` and `caption_state` outputs. The app reuses one session for both inputs. No separate `caption_encoder.onnx` is needed. Requires the v4.1-compatible iOS client. - `500m-v3/`: standard Japanese TTS and reference-audio voice cloning. - `600m-v3-vd/`: VoiceDesign model with caption conditioning. Each pack contains the split `context_kv.onnx` + `dit_step.onnx` execution path used by the Onsei iOS app, DACVAE encode/decode, speaker and text encoders, duration prediction, configuration, and tokenizer data. `manifest.json` records file sizes and SHA-256 digests used by the app to verify downloads. ## Licenses and attribution The Irodori-TTS weights, Japanese Semantic-DACVAE adaptations,...
--- license: other language: - ja pipeline_tag: text-to-speech tags: - onnx - ios - on-device - irodori-tts --- # Onsei iOS ONNX Model Packs ONNX model packs exported for the Onsei iOS client from the following upstream models: - [`Aratako/Irodori-TTS-500M-v3`](https://huggingface.co/Aratako/Irodori-TTS-500M-v3), source revision `236c1e56591279fc24e3c1bf6609fc06e48dde28` - [`Aratako/Irodori-TTS-600M-v3-VoiceDesign`](https://huggingface.co/Aratako/Irodori-TTS-600M-v3-VoiceDesign), source revision `e863a3a93e652e09afeff3e84823a206a0a60314` - [`Aratako/Semantic-DACVAE-Japanese-32dim`](https://huggingface.co/Aratako/Semantic-DACVAE-Japanese-32dim) - Tokenizer from [`llm-jp/llm-jp-3-150m`](https://huggingface.co/llm-jp/llm-jp-3-150m) These files are converted runtime artifacts, not new model weights. They are not affiliated with or endorsed by the upstream authors. ## Packs - `500m-v3/`: standard Japanese TTS and reference-audio voice cloning. - `600m-v3-vd/`: VoiceDesign model with caption conditioning. Each pack contains the split `context_kv.onnx` + `dit_step.onnx` execution path used by the Onsei iOS app, DACVAE encode/decode, speaker and text encoders, duration prediction, configuration, and tokenizer data. `manifest.json` records file sizes and SHA-256 digests used by the app to verify downloads. ## Licenses and attribution The Irodori-TTS model weights, Semantic-DACVAE model weights, and conversion code are distributed under the MIT License. The llm-jp tokenizer is distributed under Apache License 2.0. See `LICENSES/` and `THIRD_PARTY_NOTICES.md`. Each component remains subject to its own license; no relicensing is implied. ## Responsible use The upstream model cards request that users: 1. Do not clone or impersonate a person's voice without explicit consent. 2. Do not create misleading deepfakes or synthetic speech intended to spread misinformation. 3. Recognize that text-only output may coincidentally resemble a real voice. 4. Take responsibility for compliance with applicable law. The ONNX export used here does not add a SilentCipher watermark. Applications using these files should clearly disclose synthetic audio where appropriate and must enforce consent for voice cloning. ## Citation ```bibtex @misc{irodori-tts-v3, author = {Chihiro Arata}, title = {Irodori-TTS: A Flow Matching-based Text-to-Speech Model with Emoji-driven Sty...
Source context: 87 downloads · 0 likes · Pipeline text-to-speech · Repo raratu/Onsei-iOS-Models
