Accessibility improvements: show friendly names in the file dropdowns - #1280
Open
froggeric wants to merge 1 commit into
Open
Accessibility improvements: show friendly names in the file dropdowns#1280froggeric wants to merge 1 commit into
froggeric wants to merge 1 commit into
Conversation
Member
|
at the end we will be the most accessible app ever |
blaisewf
marked this pull request as ready for review
August 27, 2026 15:00
Contributor
Author
|
I wish it was that simple. Unfortunately, it is very difficult to make a web UI app accessible, like a native app would be. I am almost done for now. Next though, I need to start working on a few small fixes to make Gradio more accessibility friendly... Thanks for merging all the previous PRs. |
froggeric
added a commit
to froggeric/applio-macOS-native-app
that referenced
this pull request
Sep 2, 2026
…ries the five fork-only toasts Test-build deletion for PR7 verification (tuple-choices IAHispano#1280). The merge brings upstream IAHispano#1276, which took the five Track-B surfaces (voice_blender, plugins, realtime, model-info, tensorboard) natively — the slim patcher's anchors now miss AND re-patching would double-toast. Delete the patcher, its 5 build registrations, the fixtures suite entry (test_patch_fixtures 6->5) and the voice_blender CASES entry in test_patcher_exit_codes. Mirrors the Track A deletion b26275b; reverts cb0e3d4's code portion. This is also the pending sync work — lands on this TEST BRANCH only. (cherry picked from commit 5e634319ed55a4e9805c533df8050987b0e2eee5)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The model, index and audio dropdowns show raw relative paths. Picking a model with a screen reader is tedious and easily prone to error. For example, when listing a model named "crepe_han_window_Titan48k", the user could first have to listening to a long string such as "logs/AttentiveWindow/crepe_han_window_Titan48k/crepe_han_window_Titan48k.pth". Imagine having to go through a list of a few of those...
Gradio dropdowns accept (label, value) pairs. The label is what the user sees and hears. The value is what the handler uses. This PR modifies the file dropdowns in the inference, TTS, real-time and training tabs to use the labels (file name without extension, such as "crepe_han_window_Titan48k"). When two files have the same name, the parent folder is appended to tell them apart. The value remain unchanged, exact path, ensuring every handler and downstream function receives exactly the same data as before this PR. Sorting is unchanged.
One small existing bug was found and fixed while making this change: the "Refresh embedders" button on the training tab returned a list that set the dropdown's value, but it never refreshed. Newly added embedders didn't appear until reload. It now returns a proper update, matching what the same button does on the other tabs.
The SID, voice and device lists already read concisely and are untouched.
Verified successfully on macos with voiceover.