Improved output parsing, small refactoring of LLM client #504
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
Improvements for output parsing:
- handling of consecutive sentence terminators (.!? etc.)
- respect max number of sentences again
- improved fusing of sentences that are too short for the TTS
- narration parsing now can deal with sentences marked as speech by quotes ("")
- created clearly readable enums to indicate if a sentence is speech or narration
Small refactoring of the LLM client
- created an `AIClient` interface that is now used throughout the software instead of the `LLMClient`
- consolidated the 5 different methods to measure text by tokens into 2 (`get_count_tokens` and `is_too_long`)
- added an alternate implementation of the `AIClient` interface called `LLMTestClient` that can be initiated with specific responses to test the parsing logic
Fixed dropping of duplicate summary paragraphs in context.
Fixed several typing issues.
This reverts commit 99708ad.
art-from-the-machine
approved these changes
Feb 12, 2025
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Outdated suggestions cannot be applied.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.
Improvements for output parsing:
...)Small refactoring of the LLM client
AIClientinterface that is now used throughout the software instead of theLLMClientget_count_tokensandis_too_long)AIClientinterface calledLLMTestClientthat can be initiated with specific responses to test the parsing logicFixed dropping of duplicate summary paragraphs in context.
Fixed several typing issues.