Chatterbox-Flash-CoreML

aufklarer/Chatterbox-Flash-CoreML

Chatterbox-Flash-CoreML is a text-to-speech model from aufklarer. Use it when you need text read aloud. It is set up for coremltools. The card lists the license as mit.

Compiled Core ML export of ResembleAI/chatterbox-flash for Apple runtimes.

Downloads · 30 days

5.8K

38% of all-time downloads

All-time downloads

15.3K

Public

Repo size

2.4 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin2.4 GB · 100%

At a glance

Task: Text-to-Speech
Library: coremltools
License: mit
Model type: chatterbox_flash_coreml
Access: Public
Created: Jul 7, 2026
Updated: Jul 7, 2026
SHA: e845c9c0

Base models

ResembleAI/chatterbox-flash

Model card

From the Hugging Face model README

Chatterbox Flash CoreML

Compiled Core ML export of ResembleAI/chatterbox-flash for Apple runtimes.

This bundle contains the Chatterbox Flash T3 token generator and the S3Gen audio back-half as static-shape .mlmodelc graphs. It is intended for an application runtime that owns text tokenization, T3 denoising/sampling, reference conditioning, and graph orchestration.

Model

Component	Parameters	Format	Precision	Static shape
T3 block-diffusion token generator	532.4M	Core ML `.mlmodelc`	fp16	text_len 256, block_size 16, max_seq 1024
S3Gen audio back-half	266.0M	Core ML `.mlmodelc`	fp16	token_len 192, mel_len 384
Total	798.4M	Core ML `.mlmodelc`	fp16	24 kHz waveform output

Files

Path	Size	Description
`config.json`	4 KB	Root metadata for download tracking and runtime discovery
`t3/ConditioningEncoder.mlmodelc`	25 MB	Speaker/prompt/emotion conditioning to T3 conditioning embedding
`t3/TextPrefill.mlmodelc`	963 MB	Causal `[cond, text, start_speech]` prefix prefill and flat KV cache
`t3/BlockDecoder.mlmodelc`	1.0 GB	Full-visible Flash speech-block logits with explicit KV cache
`t3/uncond_block_prior.npy`	36 KB	Unconditional PMI prior for Flash scoring
`t3/tokenizer.json`	28 KB	Chatterbox Flash text tokenizer
`t3/config.json`	4 KB	T3 export metadata
`audio/FlowSpeakerProjector.mlmodelc`	44 KB	S3Gen reference embedding to projected speaker conditioning
`audio/FlowEncoder.mlmodelc`	79 MB	`prompt_token ++ speech_tokens` to flow `mu` and mask
`audio/FlowEstimator.mlmodelc`	141 MB	One meanflow Euler derivative step
`audio/HiFTVocoder.mlmodelc`	41 MB	Mel frames to 24 kHz waveform
`audio/audio_config.json`	4 KB	S3Gen audio export metadata

Runtime Boundary

The exported graphs cover:

T3 conditioning, text prefill, and block decoding.
S3Gen speaker projection, flow encoder, meanflow estimator step, and HiFT vocoder.

The host runtime must still provide:

text normalization and tokenization
T3 denoising loop, PMI/CFG scoring, sampling, and EOS trimming
reference waveform encoders:
- VoiceEncoder: ref.wav -> speaker_emb
- prompt speech tokenizer: ref.wav -> prompt_speech_tokens
- S3Gen reference encoder: ref.wav -> prompt_token, prompt_feat, embedding
padding/cropping and the two-step meanflow loop over (t,r) = (0,0.5), (0.5,1.0)

This means the bundle supports voice-cloning TTS when the runtime supplies reference conditioning tensors, but it is not yet a fully Core ML ref.wav -> cloned wav pipeline.

Validation

Test	Result
T3 graph roundtrip vs PyTorch wrappers	Pass at 2% relative tolerance
Audio graph roundtrip vs PyTorch wrappers	Pass for token_len 192, mel_len 384
Stitched S3Gen meanflow-to-audio roundtrip	Pass
Prompted synthesis smoke test	Pass
Whisper tiny transcript of generated wav	`Core ML speech test.`
Smoke-test WER	0.000

Core ML warnings from local export:

CPU_ONLY prediction crashed for the T3 package in local coremltools 8.3 testing. Use ALL, CPU_AND_NE, or a compiled-device runtime.
The uploaded artifact ships compiled .mlmodelc folders. The numerical parity tests were run against the source .mlpackage exports before compilation.

Usage Sketch

The runtime loads graphs from t3/ and audio/, then:

Prepare T3 reference conditioning tensors and S3Gen ref_dict tensors from a prompt wav.
Tokenize text with t3/tokenizer.json.
Run T3 prefill and block denoising/sampling to produce S3 speech tokens.
Concatenate S3Gen prompt_token and generated speech tokens.
Run audio/FlowEncoder.mlmodelc.
Build cond by copying prompt_feat.T into the mel prefix.
Run audio/FlowEstimator.mlmodelc twice for (0,0.5) and (0.5,1.0).
Crop generated mel frames after the prompt prefix.
Run audio/HiFTVocoder.mlmodelc and crop padded samples.

Source

Converted from ResembleAI/chatterbox-flash, revision 4385507288b8197e6dab8b4e6b1603328d549d9d.

Task: Text-to-Speech
Library: coremltools
Type: chatterbox_flash_coreml
License: mit
Languages: en
Created: Jul 7, 2026
Updated: Jul 7, 2026