Hugging Face
Models
Datasets
Spaces
Buckets
new
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Hardware
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
FluidInference
/
pocket-tts-coreml
like
7
Follow
Fluid Inference Community
116
Text-to-Speech
Core ML
English
tts
apple
on-device
arxiv:
2410.00037
License:
cc-by-4.0
Model card
Files
Files and versions
xet
Community
3
Copy to bucket
new
main
pocket-tts-coreml
/
v2.1
Ctrl+K
Ctrl+K
2 contributors
History:
5 commits
alexwengg
Add speaker_proj_weight (24L) + encoder_recover_pinv for voice-clone reprojection (FluidAudio #793)
1bd2078
verified
13 days ago
english
feat(v2.1/english): pocket_state multifunction MLState package (write_state/prefill/generate over shared KV state; mobius Trial 23). Loaded by FluidAudio PocketTtsModelPlacement.aneState (macOS 15+/iOS 18+); Swift-measured -14.9% e2e vs .ane, WER 0.
about 1 month ago
french_24l
Add speaker_proj_weight (24L) + encoder_recover_pinv for voice-clone reprojection (FluidAudio #793)
13 days ago
german
feat(v2.1): ANE-eligible flowlm_step_ane + cond_prefill_ane for all 9 remaining language packs (german/italian/portuguese/spanish 6L + french/german/italian/portuguese/spanish 24L). Rank-4 split-KV, scatter-free, explicit output names, no NaN-BOS protocol (mobius Trials 19/20). flowlm 100% ANE on every pack incl. 24L (1830 ops). Loaded by FluidAudio PocketTtsModelPlacement.ane.
about 1 month ago
german_24l
Add speaker_proj_weight (24L) + encoder_recover_pinv for voice-clone reprojection (FluidAudio #793)
13 days ago
italian
feat(v2.1): ANE-eligible flowlm_step_ane + cond_prefill_ane for all 9 remaining language packs (german/italian/portuguese/spanish 6L + french/german/italian/portuguese/spanish 24L). Rank-4 split-KV, scatter-free, explicit output names, no NaN-BOS protocol (mobius Trials 19/20). flowlm 100% ANE on every pack incl. 24L (1830 ops). Loaded by FluidAudio PocketTtsModelPlacement.ane.
about 1 month ago
italian_24l
Add speaker_proj_weight (24L) + encoder_recover_pinv for voice-clone reprojection (FluidAudio #793)
13 days ago
portuguese
feat(v2.1): ANE-eligible flowlm_step_ane + cond_prefill_ane for all 9 remaining language packs (german/italian/portuguese/spanish 6L + french/german/italian/portuguese/spanish 24L). Rank-4 split-KV, scatter-free, explicit output names, no NaN-BOS protocol (mobius Trials 19/20). flowlm 100% ANE on every pack incl. 24L (1830 ops). Loaded by FluidAudio PocketTtsModelPlacement.ane.
about 1 month ago
portuguese_24l
Add speaker_proj_weight (24L) + encoder_recover_pinv for voice-clone reprojection (FluidAudio #793)
13 days ago
spanish
feat(v2.1): ANE-eligible flowlm_step_ane + cond_prefill_ane for all 9 remaining language packs (german/italian/portuguese/spanish 6L + french/german/italian/portuguese/spanish 24L). Rank-4 split-KV, scatter-free, explicit output names, no NaN-BOS protocol (mobius Trials 19/20). flowlm 100% ANE on every pack incl. 24L (1830 ops). Loaded by FluidAudio PocketTtsModelPlacement.ane.
about 1 month ago
spanish_24l
Add speaker_proj_weight (24L) + encoder_recover_pinv for voice-clone reprojection (FluidAudio #793)
13 days ago