### Install DeepSeek SDK and Configure Credentials

Source: https://repo2txt.com/blog/deepseek-r1-guide.html

Install the necessary SDK and configure your API key and environment to access the DeepSeek client.

```python
# Install DeepSeek SDK
pip install deepseek-r1

# Configure credentials
import deepseek
client = deepseek.Client(
    api_key="YOUR_API_KEY",
    environment="prod" 
)
```

--------------------------------

### Install LiveKit SDK and Agents

Source: https://repo2txt.com/blog/livekit-voice-agents-with-urdu-tts.html

Install the core LiveKit Python SDK, the agents framework, and any necessary AI service plugins using pip.

```bash
pip install livekit livekit-agents livekit-plugins-openai
```

--------------------------------

### Install LiveKit Uplift Plugin

Source: https://repo2txt.com/blog/livekit-voice-agents-with-urdu-tts.html

Install the Uplift plugin using pip. Ensure your Uplift AI API key is set as an environment variable UPLIFT_AI_API_KEY.

```bash
pip install livekit-plugins-uplift
```

--------------------------------

### Start SmolVLM Local Server

Source: https://repo2txt.com/blog/smolvlm-realtime-webcam-ai-demo.html

Launches a local HTTP API for image and text processing. Use the -ngl flag for GPU acceleration. Default API address is http://localhost:8080.

```bash
llama-server -hf
# or with GPU layers enabled:
llama-server -hf -ngl 99
```

--------------------------------

### Install Dia from GitHub via pip

Source: https://repo2txt.com/blog/dia-1.6-open-source-tts-model.html

Install the latest version of the Dia model directly from its GitHub repository using pip. This method ensures you have the most up-to-date code.

```bash
pip install git+https://github.com/nari-labs/dia.git

```