[Voice-QuickFix] Start speaking filler immediately #496
Conversation
This file contains 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
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎
|
juberti
reviewed
Nov 15, 2023
|
Sure, either works. I kinda see different assistant messages as being more separate (i.e. totally separate paragraphs even, otherwise why separate them) but I don't have a strong opinion here. I'll change it to a space. |
petersalas
approved these changes
Nov 15, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
I noticed that the filler sentences (or interjections as Peter calls them, which I'm not a total fan of), don't start until after a long pause and seemed to only start after the corpus response was ready, which defeats the purpose. The issue seemed to have been due to the filler not being recognized as a separate sentence which meant TTS would not be called.
This PR adds a newline between separate
AssistantMessageresponses from Fixie to solve this issue for Fixie calls.Note that I did not investigate what happens in the other cases (LLM not from Fixie).
Logs before: notice no chunk is being requested from TTS for a long time.
Logs after: TTS being called before actual content is ready, though it's still not easy to understand what's happening underneath when simply listening to it (there's no gap between filler TTS end and start of actual content).
