videopython-chatterbox 0.1.7.post1

Chatterbox TTS (Resemble AI) with corrected dependency metadata. Upstream source, one line changed.

videopython-chatterbox

Chatterbox TTS by Resemble AI, republished with corrected dependency metadata.

The Python source is unchanged from upstream chatterbox-tts 0.1.7 except for a single line (the self-version lookup in chatterbox/__init__.py, which must name this distribution or the package fails to import). Everything else that differs is dependency metadata in pyproject.toml.

pip install videopython-chatterbox

The import name is unchanged, so this is a drop-in replacement:

from chatterbox.mtl_tts import ChatterboxMultilingualTTS

Conflict warning: this package and chatterbox-tts both install a top-level chatterbox package. Install one or the other, never both.

Why this fork exists

Upstream chatterbox-tts 0.1.7 pins diffusers==0.29.0, torch==2.6.0, torchaudio==2.6.0, transformers==5.2.0 and safetensors==0.5.3. Those == pins make it unsatisfiable alongside any modern ML stack — for example pyannote-audio>=4.0 requires torch>=2.8.0, so the two cannot coexist:

$ pip install chatterbox-tts diffusers>=0.35.0
  × No solution found: chatterbox-tts==0.1.7 depends on diffusers==0.29.0

The pins are not load-bearing. Chatterbox touches diffusers in exactly one place — its vendored Matcha-TTS decoder — importing 9 symbols (GEGLU, GELU, AdaLayerNorm, AdaLayerNormZero, ApproximateGELU, Attention, LoRACompatibleLinear, maybe_allow_in_graph, get_activation). All of them still exist and behave identically in diffusers 0.39.

Verification

Real synthesis and voice cloning were run on torch 2.13.0 / diffusers 0.39.0 / transformers 5.14.1 / numpy 2.4.6, then repeated under upstream's pinned transformers==5.2.0, with fixed seeds:

seed	samples	identical	max abs diff
1234	91200	yes	0.0
1235	86400	yes	0.0
1236	144000	yes	0.0

Output is bit-identical. Generations were also transcribed with Whisper to confirm the audio matches the requested text, rather than merely being non-silent.

Metadata changes

dependency	upstream 0.1.7	here	why
numpy	<2.0.0 (py<3.13)	>=1.24.0	verified on 2.4.6
librosa	==0.11.0	>=0.11.0	0.11.0 is still latest
torch	==2.6.0	>=2.6.0	verified on 2.13.0
torchaudio	==2.6.0	>=2.6.0	follows torch
transformers	==5.2.0	>=5.2.0	verified bit-identical on 5.14.1
diffusers	==0.29.0	>=0.29.0,<1.0	verified on 0.39.0; see cap below
conformer	==0.3.2	>=0.3.2	0.3.2 is still latest
safetensors	==0.5.3	>=0.5.3	verified on 0.8.0
pykakasi	==2.3.0	>=2.3.0	2.3.0 is still latest
gradio	==6.8.0	removed	not imported anywhere in the package
setuptools	undeclared	<81	see below

Two of these are deliberate constraints rather than relaxations:

diffusers<1.0 — LoRACompatibleLinear is deprecated with removal scheduled for diffusers 1.0.0. It is the one symbol that will genuinely break, so the cap is real.

setuptools<81 — resemble-perth declares no dependencies at all and imports pkg_resources, which setuptools removed in 82+. Its __init__.py swallows the resulting ImportError and sets PerthImplicitWatermarker = None, so chatterbox fails later at model load with a confusing TypeError: 'NoneType' object is not callable. Upstream is affected by this too; it is simply undeclared there.

Relationship to upstream

No functional changes are intended, ever. If Resemble AI relaxes their pins, this fork becomes unnecessary and should be abandoned in favour of chatterbox-tts.

To rebase onto a new upstream release: replace src/chatterbox/ wholesale, re-apply the one-line __version__ change in chatterbox/__init__.py, verify every other file is byte-identical to the new sdist, bump version to <upstream>.postN, and re-check the dependency table above.

License

MIT, Copyright (c) 2025 Resemble AI — preserved verbatim in LICENSE. See NOTICE for the fork's provenance statement. Model weights are not redistributed; they are fetched at runtime from the Hugging Face Hub.