# Whisper AI
[[transcribing]]
![[01 Unsorted/Recording 20240518181844.webm]]
[[count not be found - embedded audio issue in obsisian]] #ObsidianBug
## Issue (solved)
- Solved - it was on chromebook and I needed to allow linux to use microphone
- 2024-05-18 - not working, responding with "you",
- Options... [[you do not need to fix this now]]
- Guesses.
- issue with api (so just wait)
- issue with my billing (check)
- Issue with the plugin
- Issue with wifi? X
- Issue with speaker x
- Issue with recording.. maybe
- works in chrome
- does not work in obsidian
Base whisper can't differentiate speakers...
Azue might be able to... this is ([[diarization]])
https://learn.microsoft.com/en-us/azure/ai-services/speech-service/whisper-overview
Google Cloud - text to speech also does it.
## Misc
- I want this in [[Obsidian]] and there is a plugin that allows for it.
- The cost is about $0.01/minute. (they say 0.006/minute)
- 1440 are the minutes in a day so ~10-$14/day.
2024-03-14 -
[[Whisper AI]] (trying it locally 2023-11-10)
- https://www.assemblyai.com/blog/how-to-run-openais-whisper-speech-recognition-model/
- THESE INSTRUCTION DID NOT WORK