This flat out cannot be installed following the directions.

[spike4379]

Using python 3.11.9

Do the following on a clean system.

git clone https://github.com/resemble-ai/chatterbox/
cd chatterbox
python -m venv venv
call venv/scripts/activate
pip install -U pip setuptools wheel
pip install "numpy>=1.24,<1.26"
pip install --no-build-isolation "pkuseg==0.0.25"

It will fail every time.

However if you remove the pkuseg from the pyproject.toml and continue using

pip install -e .

When running with the command set

echo
call venv/scripts/activate
python gradio_tts_app.py --port 7861 --share=false
pause

you will get the error
\chatterbox\venv\Lib\site-packages\perth\perth_net_init_.py:1: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81.
from pkg_resources import resource_filename
Traceback (most recent call last):
File "F:\ChatGPT\chatterbox\gradio_tts_app.py", line 5, in
from chatterbox.tts import ChatterboxTTS
File "F:\ChatGPT\chatterbox\src\chatterbox_init_.py", line 9, in
from .tts import ChatterboxTTS
File "F:\ChatGPT\chatterbox\src\chatterbox\tts.py", line 12, in
from .models.s3tokenizer import S3_SR, drop_invalid_tokens
File "F:\ChatGPT\chatterbox\src\chatterbox\models\s3tokenizer_init_.py", line 1, in
from .s3tokenizer import (
File "F:\ChatGPT\chatterbox\src\chatterbox\models\s3tokenizer\s3tokenizer.py", line 7, in
from s3tokenizer.utils import padding
ModuleNotFoundError: No module named 's3tokenizer'
Press any key to continue . . .

I have also done a pip freeze to list all packages inside the venv

aiofiles==24.1.0
annotated-types==0.7.0
anyio==4.10.0
audioread==3.0.1
Brotli==1.1.0
catalogue==2.0.10
certifi==2025.8.3
cffi==1.17.1
cfgv==3.4.0
charset-normalizer==3.4.3
-e git+https://github.com/resemble-ai/chatterbox@1b5ae50585c8fa4d30ca6a7b304d787b4f802a42#egg=chatterbox_tts
click==8.2.1
colorama==0.4.6
conformer==0.3.2
decorator==5.2.1
Deprecated==1.2.18
diffusers==0.29.0
distlib==0.4.0
einops==0.8.1
fastapi==0.116.1
ffmpy==0.6.1
filelock==3.19.1
fsspec==2025.9.0
gradio==5.44.1
gradio_client==1.12.1
groovy==0.1.2
h11==0.16.0
httpcore==1.0.9
httpx==0.28.1
huggingface-hub==0.34.4
identify==2.6.13
idna==3.10
importlib_metadata==8.7.0
jaconv==0.4.0
Jinja2==3.1.6
joblib==1.5.2
lazy_loader==0.4
librosa==0.11.0
llvmlite==0.44.0
markdown-it-py==4.0.0
MarkupSafe==3.0.2
mdurl==0.1.2
ml_dtypes==0.5.3
mpmath==1.3.0
msgpack==1.1.1
networkx==3.5
nodeenv==1.9.1
numba==0.61.2
numpy==1.25.2
onnx==1.19.0
orjson==3.11.3
packaging==25.0
pandas==2.3.2
pillow==11.3.0
platformdirs==4.4.0
pooch==1.8.2
pre_commit==4.3.0
protobuf==6.32.0
pycparser==2.22
pydantic==2.11.7
pydantic_core==2.33.2
pydub==0.25.1
Pygments==2.19.2
pykakasi==2.3.0
python-dateutil==2.9.0.post0
python-multipart==0.0.20
pytz==2025.2
PyYAML==6.0.2
regex==2025.9.1
requests==2.32.5
resemble-perth==1.0.1
rich==14.1.0
ruff==0.12.12
s3tokenizer==0.2.0
safehttpx==0.1.6
safetensors==0.5.3
scikit-learn==1.7.1
scipy==1.16.1
semantic-version==2.10.0
shellingham==1.5.4
six==1.17.0
sniffio==1.3.1
soundfile==0.13.1
soxr==0.5.0.post1
spacy_pkuseg==1.0.1
srsly==2.5.1
starlette==0.47.3
sympy==1.13.1
threadpoolctl==3.6.0
tokenizers==0.20.3
tomlkit==0.13.3
torch==2.6.0
torchaudio==2.6.0
tqdm==4.67.1
transformers==4.46.3
typer==0.17.4
typing-inspection==0.4.1
typing_extensions==4.15.0
tzdata==2025.2
urllib3==2.5.0
uvicorn==0.35.0
virtualenv==20.34.0
websockets==15.0.1
wrapt==1.17.3
zipp==3.23.0

I have tried this with conda as well and got the same result.

[spike4379]

To update, also using the process using conda
conda create -yn chatterbox python=3.11
conda activate chatterbox
pip install -U pip setuptools wheel
pip install chatterbox-tts
Collecting chatterbox-tts
Downloading chatterbox_tts-0.1.4-py3-none-any.whl.metadata (9.2 kB)
Collecting numpy<1.26.0,>=1.24.0 (from chatterbox-tts)
Downloading numpy-1.25.2-cp311-cp311-win_amd64.whl.metadata (5.7 kB)
Collecting librosa==0.11.0 (from chatterbox-tts)
Downloading librosa-0.11.0-py3-none-any.whl.metadata (8.7 kB)
Collecting s3tokenizer (from chatterbox-tts)
Downloading s3tokenizer-0.2.0.tar.gz (225 kB)
Preparing metadata (setup.py) ... done
Collecting torch==2.6.0 (from chatterbox-tts)
Downloading torch-2.6.0-cp311-cp311-win_amd64.whl.metadata (28 kB)
Collecting torchaudio==2.6.0 (from chatterbox-tts)
Downloading torchaudio-2.6.0-cp311-cp311-win_amd64.whl.metadata (6.7 kB)
Collecting transformers==4.46.3 (from chatterbox-tts)
Downloading transformers-4.46.3-py3-none-any.whl.metadata (44 kB)
Collecting diffusers==0.29.0 (from chatterbox-tts)
Downloading diffusers-0.29.0-py3-none-any.whl.metadata (19 kB)
Collecting resemble-perth==1.0.1 (from chatterbox-tts)
Downloading resemble_perth-1.0.1-py3-none-any.whl.metadata (4.8 kB)
Collecting conformer==0.3.2 (from chatterbox-tts)
Downloading conformer-0.3.2-py3-none-any.whl.metadata (631 bytes)
Collecting safetensors==0.5.3 (from chatterbox-tts)
Downloading safetensors-0.5.3-cp38-abi3-win_amd64.whl.metadata (3.9 kB)
Collecting pkuseg==0.0.25 (from chatterbox-tts)
Downloading pkuseg-0.0.25.tar.gz (48.8 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 48.8/48.8 MB 1.2 MB/s 0:00:40
Preparing metadata (setup.py) ... error
error: subprocess-exited-with-error

× python setup.py egg_info did not run successfully.
│ exit code: 1
╰─> [6 lines of output]
Traceback (most recent call last):
File "", line 2, in
File "", line 35, in
File "C:\Users\user1\AppData\Local\Temp\pip-install-q8nu1r98\pkuseg_9604ba474a064490b09d4734ae5b8d4f\setup.py", line 5, in
import numpy as np
ModuleNotFoundError: No module named 'numpy'
[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
error: metadata-generation-failed

× Encountered error while generating package metadata.
╰─> See above for output.

note: This is an issue with the package mentioned above, not pip.
hint: See above for details.

Both listed methods to install and run this are not working for a normal user.

[nfranke]

I finally got it to work, although I'm running from the src directory, i.e. I didn't install the package.

conda create -yn chatterbox python=3.11
conda activate chatterbox
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu128
pip install "numpy>=1.24,<1.26"
pip install librosa huggingface_hub safetensors transformers einops s3tokenizer conformer diffusers peft resemble-perth 

Then I modified example_vc.py to add this below the imports:

import sys
import os

sys.path.append(os.path.join(os.path.dirname(__file__), "src"))

Then I can run:

python example_vc.py

without errors.

[spike4379]

I finally got it to work, although I'm running from the src directory, i.e. I didn't install the package.

conda create -yn chatterbox python=3.11
conda activate chatterbox
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu128
pip install "numpy>=1.24,<1.26"
pip install librosa huggingface_hub safetensors transformers einops s3tokenizer conformer diffusers peft resemble-perth 

Then I modified example_vc.py to add this below the imports:

import sys
import os

sys.path.append(os.path.join(os.path.dirname(__file__), "src"))

Then I can run:

python example_vc.py

without errors.

Nice! Have you been able to get the gradio_tts_app.py to run?

[tetelias]

To update, also using the process using conda conda create -yn chatterbox python=3.11 conda activate chatterbox pip install -U pip setuptools wheel pip install chatterbox-tts Collecting chatterbox-tts Downloading chatterbox_tts-0.1.4-py3-none-any.whl.metadata (9.2 kB) Collecting numpy<1.26.0,>=1.24.0 (from chatterbox-tts) Downloading numpy-1.25.2-cp311-cp311-win_amd64.whl.metadata (5.7 kB) Collecting librosa==0.11.0 (from chatterbox-tts) Downloading librosa-0.11.0-py3-none-any.whl.metadata (8.7 kB) Collecting s3tokenizer (from chatterbox-tts) Downloading s3tokenizer-0.2.0.tar.gz (225 kB) Preparing metadata (setup.py) ... done Collecting torch==2.6.0 (from chatterbox-tts) Downloading torch-2.6.0-cp311-cp311-win_amd64.whl.metadata (28 kB) Collecting torchaudio==2.6.0 (from chatterbox-tts) Downloading torchaudio-2.6.0-cp311-cp311-win_amd64.whl.metadata (6.7 kB) Collecting transformers==4.46.3 (from chatterbox-tts) Downloading transformers-4.46.3-py3-none-any.whl.metadata (44 kB) Collecting diffusers==0.29.0 (from chatterbox-tts) Downloading diffusers-0.29.0-py3-none-any.whl.metadata (19 kB) Collecting resemble-perth==1.0.1 (from chatterbox-tts) Downloading resemble_perth-1.0.1-py3-none-any.whl.metadata (4.8 kB) Collecting conformer==0.3.2 (from chatterbox-tts) Downloading conformer-0.3.2-py3-none-any.whl.metadata (631 bytes) Collecting safetensors==0.5.3 (from chatterbox-tts) Downloading safetensors-0.5.3-cp38-abi3-win_amd64.whl.metadata (3.9 kB) Collecting pkuseg==0.0.25 (from chatterbox-tts) Downloading pkuseg-0.0.25.tar.gz (48.8 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 48.8/48.8 MB 1.2 MB/s 0:00:40 Preparing metadata (setup.py) ... error error: subprocess-exited-with-error

× python setup.py egg_info did not run successfully. │ exit code: 1 ╰─> [6 lines of output] Traceback (most recent call last): File "", line 2, in File "", line 35, in File "C:\Users\user1\AppData\Local\Temp\pip-install-q8nu1r98\pkuseg_9604ba474a064490b09d4734ae5b8d4f\setup.py", line 5, in import numpy as np ModuleNotFoundError: No module named 'numpy' [end of output]

note: This error originates from a subprocess, and is likely not a problem with pip. error: metadata-generation-failed

× Encountered error while generating package metadata. ╰─> See above for output.

note: This is an issue with the package mentioned above, not pip. hint: See above for details.

Both listed methods to install and run this are not working for a normal user.

It does say

No module named 'numpy'

I ran

conda install numpy==1.25.2"

It will then ask for Cython, so you also need to run

conda install Cython

and installation finished successfully.

[nfranke]

I installed numpy as part of my script. I didn't run the gradio script and I didn't install cython. Maybe that's for installation, which I didn't do. The "example_tts.py" script worked, though.

[shashank404error]

You need to stop using watermark in src/chatterbox/tts.py
Also you may not need to download 'perth' lib provided by them if you stop using watermark.

[umbraprior]

Ensure you are using Python 3.11 or 3.12 then run the following:

python -m pip install --upgrade pip setuptools wheel

pip install pkuseg==0.0.25 --no-build-isolation

pip install -e .