Downloads · 30 days
840
7% of all-time downloads
FluidInference/pocket-tts-coreml
pocket-tts-coreml is a text-to-speech model from FluidInference. Use it when you need text read aloud. It is set up for coreml. The card lists the license as cc-by-4.0.
CoreML conversion of kyutai/pocket-tts for on-device inference on Apple platforms.
Downloads · 30 days
840
7% of all-time downloads
All-time downloads
11.7K
Public
Repo size
48 GB
Likes
8
Public
Click a slice to open those files.
.bin73.1 GB · 89%
From the Hugging Face model README
CoreML conversion of kyutai/pocket-tts for on-device inference on Apple platforms.
| Model | Description | Size |
|---|---|---|
| cond_step | KV cache prefill (voice + text conditioning) | ~200MB |
| flowlm_step | Autoregressive generation (transformer_out + EOS) | ~200MB |
| flow_decoder | Flow matching denoiser (8 Euler steps per frame) | ~190MB |
| mimi_decoder | Streaming audio codec (1920 samples per frame) | ~11MB |
4 pre-encoded voices in constants_bin/:
alba (default), azelma, cosette, javertVoice cloning is supported: the Mimi encoders are included (mimi_encoder*.mlmodelc at the root for English, v2.1/<lang>/mimi_encoderv3.mlmodelc per language pack). In FluidAudio, use PocketTtsManager.cloneVoice() or fluidaudiocli tts "…" --backend pocket --clone-voice speaker.wav.
import FluidAudioTTS
let manager = PocketTtsManager()
try await manager.initialize()
let audio = try await manager.synthesize(text: "Hello, world!")
See https://github.com/FluidInference/FluidAudio for the full Swift framework.
License
CC-BY-4.0, inherited from https://huggingface.co/kyutai/pocket-tts. Attribution to Kyutai is required.
References
- https://huggingface.co/kyutai/pocket-tts
- https://arxiv.org/abs/2410.00037
- https://github.com/FluidInference/FluidAudio