It is just doing some stuff intelligently.
Project description
🦈 Shirley
It is just doing some stuff intelligently. It has the following features:
- Chat, powered by Qwen/Qwen-VL-Chat (通义千问).
- Text-To-Speech, powered by Azure AI Speech.
Installation
pip install shirley
Or, copy shirley
directory to your working directory.
[Chat] Setup Qwen-VL-Chat model (local)
Add Qwen/Qwen-VL-Chat as a git submodule:
git submodule add https://huggingface.co/Qwen/Qwen-VL-Chat.git models/qwen_vl_chat
git submodule update --init --recursive models/qwen_vl_chat
git submodule update --remote --merge models/qwen_vl_chat
This downloads the model repository from Hugging Face and saves it in models/qwen_vl_chat
.
Note: You can skip this if you want to use remote model.
[Text-To-Speech] Setup Azure AI Speech
Create AI Speech service in Azure portal.
Set environment variables:
export SPEECH_KEY=your-key
export SPEECH_REGION=your-region
Set environment variables on Windows:
setx SPEECH_KEY your-key
setx SPEECH_REGION your-region
Quicktour
# webui.py
import gradio as gr
import shirley as sh
with gr.Blocks() as blocks:
sh.interfaces.Header()
with gr.Tab('Chat'):
sh.interfaces.Chat()
with gr.Tab('Text-To-Speech'):
sh.interfaces.TextToSpeech()
sh.interfaces.Footer()
blocks.queue().launch()
Chat interface uses local model by default. If you want to use remote one, add options:
...
sh.interfaces.Chat(
options=ChatInterfaceOptions(
client=ChatClientOptions(local=False),
),
)
...
Note: I made an example webui.py. You can use that to start with.
Running
pip install -r requirements.txt
python webui.py
Note: I recommend using poetry
to manage dependencies and run Python. See DEVELOPMENT.md for more details.
Development
See DEVELOPMENT.md.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
File details
Details for the file shirley-0.1.19.tar.gz
.
File metadata
- Download URL: shirley-0.1.19.tar.gz
- Upload date:
- Size: 12.1 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.9.19
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 47fefb75c121cb97b91a77f32fc06a1a5c72bbc71fec9ecc20744bda7249f492 |
|
MD5 | 3b3572faff400b3520aaff26cfa2f829 |
|
BLAKE2b-256 | f1fadbdbae6a2fbdcf022906369920766811171aaf25588fa785a4cd8a52c532 |
File details
Details for the file shirley-0.1.19-py3-none-any.whl
.
File metadata
- Download URL: shirley-0.1.19-py3-none-any.whl
- Upload date:
- Size: 16.9 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.9.19
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 3c6549901f684746b6fe8056930419e0692e0790efeb87863feb367a9c260997 |
|
MD5 | f0c8565a229545dbace0c29145314831 |
|
BLAKE2b-256 | 0baa2095769d701d1a87cc324bc16439bb7eef044c16dc9c4c7bcb6b0422f65e |