Scribe 2.3.1 - Feature - Enables the user to pause a recording

Scribe 2.3.1 - Feature: Enables Pause

This has been a long standing request, it's nice to finally have it in.

What's Changed

• 51 pauze functionality request by @Mikodin in #98

Full Changelog: 2.3.0...2.3.1

Scribe 2.3.0 - Bug Fix - LLM summarization completion cancelling new recording

Scribe 2.3.0 - Bug Fix - LLM summarization completion cancelling new recording

Blocks the user from beginning a new recording when there is LLM processing happening in the background.

This is the simplest solution to this problem.
The reality is that it would beautiful to be able to simply start more recordings. In order to enable this, we would need to rearchitecture the app, and the changes would be much larger than I would like in order to include a queuing system of sorts.

And so, this will do it :)

What's Changed

• 48 llm summarization complete cancels the recording by @Mikodin in #97

Full Changelog: 2.2.9...2.3.0

Scribe 2.2.9 - Bug Fix Appending Audio Overwriting Audio frontmatter instead of appending

Scribe 2.2.9 - Bug Fix Appending Audio Overwriting Audio frontmatter instead of appending

This fix has two components to it

1. We are properly appending to the audio frontmatter instead of overwriting
2. We are writing the frontmatter immediately after audio file creation. This enables us to not have to dig for the audio file if our transcription request fails.

What's Changed

• 79 audio property is incorrect when using save audio append to current multiple recordings by @Mikodin in #96

Full Changelog: 2.2.8...2.2.9

Scribe 2.2.8 Shows an indicator when recording from "Start recording"

Shows an indicator when recording from "Start recording"

#82

What's Changed

• Adds the recording notice for a visual indicator while recording by @Mikodin in #95

Full Changelog: 2.2.7...2.2.8

Scribe 2.2.7 - Bug Fix AssemblyAI speech_models being sometimes empty

Scribe 2.2.7 - Bug Fix AssemblyAI speech_models being sometimes empty

Fixes issue reported here #94
Updates the assemblyAI version and codifies the model being chosen.

What's Changed

• Fixes assemblyAi speech_models sometimes being empty by @Mikodin in #94

Full Changelog: 2.2.6...2.2.7

Scribe 2.2.6 - Bug fix of modal width in Obsidian Beta

Minor bug fix that improves UI on Obsidian Beta mobile greatly...without overriding Obsidians default modal style (like in 2.2.5 - whoops)

Full Changelog: 2.2.5...2.2.6

Scribe 2.2.5 - Bug fix of modal width in Obsidian Beta

Scribe 2.2.5 - Bug fix of modal width in Obsidian Beta

Minor bug fix that improves UI on Obsidian Beta mobile greatly

Full Changelog: 2.2.4...2.2.5

Scribe 2.2.4 - Bug Fixes & Dep Updates - Fully fix openAI issues

Scribe 2.2.4

This release is solid, I tested in a wide range of scenarios, everything should be good to go. This fixes the OpenAI Whisper issue, turns out the fix-webm-duration package started to break recordings that were over 30 seconds by making audioContext.decodeAudio throw errors. This was reported in #84

I did a few things to clean up

1. Removed fix-webm-duration, this was the crux of the issue. The duration was appearing fine across the board for me on MacOS and iOS
2. Implemented and used standardized-audio-context package to remove inconsistencies (hopefully) through Webkit and Chrome, etc
3. Updated all dependencies
4. Removed MP3 options. Sorry @tggo , it wasn't working in any scenarios (ios or macos) for me and it being a desktop only thing anyway (because of ffmpeg and the webworkers), I decided to ditch it.

What's Changed

• 84 fix chunking, mime type encoding, removes mp3 support, standardize audio, update all deps by @Mikodin in #89

Full Changelog: 2.2.3...2.2.4

Scribe 2.2.3 - Bug Fix OpenAi for transcription now works

Scribe 2.2.3 Fixes decoding failed issue with OpenAI for transcription

It appears that something changed in Safari / Webkit in how decodeAudioData works. I am not sure what it is exactly, I wasn't able to get it to work in any reliable way. For now, when this error happens, we are just passing the entire file to OpenAI

This will create a known issue with file sizes larger than 25MB not working. But it's better than any recording longer than 30 seconds not working.

I need to rework file splitting. Things have gotten wonky in how files and the audiobuffer are being dealt with, I need to unify it.

What's Changed

• Update README.md by @bdwelle in #87
• 84 garbage decoding always fails take2 by @Mikodin in #88

New Contributors

• @bdwelle made their first contribution in #87

Full Changelog: 2.2.2...2.2.3

Scribe 2.2.2 Revert Bug Fix - Issue persists and may break other users experience

Reverts releases/2.2.1

What's Changed

• Revert "84 garbage decoding always fails" - issue is still happening, I went on the wrong trail by @Mikodin in #86

Full Changelog: 2.2.1...2.2.2