https://github.com/xzeldon/whisper-api-server-legacy
API server that makes transcription with the OpenAI Whisper models.
https://github.com/xzeldon/whisper-api-server-legacy
Last synced: 2 months ago
JSON representation
API server that makes transcription with the OpenAI Whisper models.
- Host: GitHub
- URL: https://github.com/xzeldon/whisper-api-server-legacy
- Owner: xzeldon
- License: mit
- Created: 2023-08-14T23:45:04.000Z (almost 2 years ago)
- Default Branch: master
- Last Pushed: 2023-10-03T22:15:08.000Z (over 1 year ago)
- Last Synced: 2025-01-18T09:33:58.969Z (4 months ago)
- Language: Python
- Homepage:
- Size: 5.86 KB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# whisper-api-server
### ⚠️ Take a look at the new version of Whisper API Server rewritten in Go! You can find it at [here](https://github.com/xzeldon/whisper-api-server). This updated version eliminates the need for approximately 6GB of Python runtime dependencies.
API server that makes transcription with the OpenAI Whisper models.
```curl
curl http://localhost:8000/v1/audio/transcriptions \
-H "Content-Type: multipart/form-data" \
-F file="@/path/to/file/audio.mp3" \
-F model="whisper-1"
```
```json
{
"text": "Imagine the wildest idea that you've ever had, and you're curious about how it might scale to something that's a 100, a 1,000 times bigger. This is a place where you can get to do that."
}
```
# About
I made this simple app for [Obsidian voice recognotion](https://github.com/nikdanilov/whisper-obsidian-plugin).
This is working pretty well this usecase. Untested on anything other than Windows.
# Usage
To use this with [Obsidian voice recognotion plugin](https://github.com/nikdanilov/whisper-obsidian-plugin) go to plugin's settings and set
- API KEY = `sk-1`
- API URL = `http://localhost:8000/v1/audio/transcriptions`
- Model = `whisper-1`
An example of using via API is [above](#whisper-api-server).
# Credits
- [This gist](https://gist.github.com/gavrilov/4537a569b7fa8e20e64a199e924d458a) as reference. But I found this implementation a bit slow, especially on every first run (about 30 seconds).