@@ -92,32 +92,60 @@ To set up the Whisper Transcriber Telegram Bot, follow these steps:
9292
9393---
9494
95- 1 . To enable GPU processing inside Docker files, install the NVIDIA Container Toolkit in i.e. Ubuntu ** on the host machine** with:
95+ To enable GPU processing inside Docker files, install the NVIDIA Container Toolkit in i.e. Ubuntu ** (on the host machine)** with these steps:
96+
97+ 1 . ** Add NVIDIA GPG Key and Repository** :
98+ Use the following commands to configure the repository securely with the GPG key:
9699
97100 ``` bash
98- distribution=$( . /etc/os-release; echo $ID$VERSION_ID ) && \
99- curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add - && \
100- curl -s -L https://nvidia.github.io/nvidia-docker/$distribution /nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list
101+ curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey | sudo gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg &&
102+ curl -s -L https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list | \
103+ sed ' s#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://#g' | \
104+ sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list
105+ ```
101106
107+ 2 . ** Update the Package List** :
108+ Run the following to refresh your package list:
109+ ``` bash
102110 sudo apt-get update
103- sudo apt-get install -y nvidia-docker2
104111 ```
105112
106- 2 . Restart Docker: After installing the NVIDIA Docker toolkit, restart the Docker daemon:
113+ 3 . ** Install the NVIDIA Container Toolkit** :
114+ Install the toolkit using:
115+ ``` bash
116+ sudo apt-get install -y nvidia-container-toolkit
117+ ```
118+
119+ 4 . ** Configure Docker to Use NVIDIA Runtime** :
120+ Configure the NVIDIA runtime for Docker:
121+ ``` bash
122+ sudo nvidia-ctk runtime configure --runtime=docker
107123 ```
124+
125+ 5 . ** Restart Docker** :
126+ Restart the Docker service to apply the changes:
127+ ``` bash
108128 sudo systemctl restart docker
109129 ```
110130
131+ 6 . ** Test the Setup** :
132+ You can verify if the setup is working correctly by running a base CUDA container:
133+ ``` bash
134+ sudo docker run --rm --runtime=nvidia --gpus all nvidia/cuda:11.6.2-base-ubuntu20.04 nvidia-smi
135+ ```
136+
137+ If everything is set up correctly, you should see your GPUs listed.
138+
111139---
112- #### Option 1: Pull the prebuilt image from GHCR
140+ #### Dockerfile Install Option 1: Pull the prebuilt image from GHCR
113141
114142Just grab the latest pre-built version with:
115143
116144 ``` bash
117145 docker pull ghcr.io/flyingfathead/whisper-transcriber-telegram-bot:latest
118146 ```
119147---
120- #### Option 2: Build the Docker image yourself
148+ #### Dockerfile Install Option 2: Build the Docker image yourself
121149
122150If there's something wrong with GHCR's prebuilt image, you can also build the Docker image yourself.
123151
0 commit comments