cjones/WhisperLive
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

update tensorrt docker & readme

Browse Source
This commit is contained in:
makaveli10
2024-01-19 07:30:50 -05:00
parent b955e63dc1
commit e3084b34cb
2 changed files with 5 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files
TensorRT_whisper.md
+3 -3
View File
@@ -12,7 +12,7 @@ cd WhisperLive
``` ```
- Build the TensorRT-LLM docker image - Build the TensorRT-LLM docker image
```bash ```bash
docker build --file docker/Dockerfile.tensorrt --tag tensorrt_llm/devel:latest . docker build --file docker/Dockerfile.tensorrt --tag tensorrt_llm:latest .
``` ```
**NOTE**: This could take some time. **NOTE**: This could take some time.
- Next, we run the docker image and mount WhisperLive repo to the containers `/home` directory. - Next, we run the docker image and mount WhisperLive repo to the containers `/home` directory.
@@ -20,7 +20,7 @@ docker build --file docker/Dockerfile.tensorrt --tag tensorrt_llm/devel:latest .
docker run -it --gpus all --shm-size=8g \ docker run -it --gpus all --shm-size=8g \
--ipc=host --ulimit memlock=-1 --ulimit stack=67108864 \ --ipc=host --ulimit memlock=-1 --ulimit stack=67108864 \
-v /path/to/WhisperLive:/home/WhisperLive \ -v /path/to/WhisperLive:/home/WhisperLive \
tensorrt_llm/devel:latest tensorrt_llm:latest
``` ```
- Once inside the docker container, make sure to test the installation. - Once inside the docker container, make sure to test the installation.
```bash ```bash
@@ -31,7 +31,7 @@ python -c "import torch; import tensorrt; import tensorrt_llm"
**NOTE**: Uncomment and update library paths if imports fail. **NOTE**: Uncomment and update library paths if imports fail.
## Whisper TensorRT Engine ## Whisper TensorRT Engine
- For this demo we build `small.en` and `small` multilingual TensorRT engine. The script logs the path of the directory with Whisper TensorRT engine. We need the model_path to run the server. - We build `small.en` and `small` multilingual TensorRT engine. The script logs the path of the directory with Whisper TensorRT engine. We need the model_path to run the server.
```bash ```bash
# convert small.en # convert small.en
bash build_whisper_tensorrt /path/to/TensorRT-LLM/examples small.en bash build_whisper_tensorrt /path/to/TensorRT-LLM/examples small.en
docker/Dockerfile.tensorrt
+2 -2
View File
@@ -5,8 +5,8 @@ ARG CUDA_ARCHITECTURES="89-real"
FROM ${BASE_IMAGE}:${BASE_TAG} as base FROM ${BASE_IMAGE}:${BASE_TAG} as base
WORKDIR /home WORKDIR /home
COPY script/install_trt_llm.sh install_trt_llm.sh COPY scripts/install_tensorrt_llm.sh install_tensorrt_llm.sh
RUN bash install_trt_llm.sh ${CUDA_ARCHITECTURES} && rm install_trt_llm.sh RUN bash install_tensorrt_llm.sh ${CUDA_ARCHITECTURES} && rm install_tensorrt_llm.sh
COPY scripts/setup.sh /home COPY scripts/setup.sh /home
RUN apt update && bash setup.sh && rm setup.sh RUN apt update && bash setup.sh && rm setup.sh