feat(stage-ui): better tts input chunking #306
Conversation
✅ Deploy Preview for airi-docs ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
✅ Deploy Preview for airi-vtuber ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
There was a problem hiding this comment.
Summary of Changes
Hello @sumimakito, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!
This pull request significantly enhances the Text-to-Speech (TTS) capabilities within the stage-ui package by introducing a sophisticated text chunking mechanism. The primary goal is to optimize the streaming of text to speech providers, leading to a faster initial response time for spoken output without compromising the coherence of sentences. This is achieved through a new utility function that intelligently segments input text based on punctuation, word counts, and a configurable 'boost' for early chunks. To facilitate testing and understanding of this new logic, a dedicated streaming playground has also been integrated into the UI.
Highlights
- Improved TTS Input Chunking: A new generator function
chunkTTSInputhas been introduced inpackages/stage-ui/src/utils/tts.tsto intelligently pre-process text into smaller, optimized chunks for Text-to-Speech (TTS) synthesis. This aims to reduce the initial wait time for speech generation while maintaining natural sentence flow by preferring complete sentences and respecting word limits. - New TTS Streaming Playground: A dedicated
SpeechStreamingPlaygroundcomponent has been added within the existing Voice Playground. This playground allows developers to easily test and visualize how text is chunked and streamed for TTS, providing insights into the new chunking logic and its effects. - Refactored Message Content Processing: The
useMessageContentQueuecomposable has been updated to leverage the newchunkTTSInpututility, replacing its previous basic sentence-splitting logic. This ensures that all TTS processing benefits from the improved, more robust chunking strategy.
Using Gemini Code Assist
The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.
Invoking Gemini
You can request assistance from Gemini at any point in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.
| Feature | Command | Description |
|---|---|---|
| Code Review | /gemini review |
Performs a code review for the current pull request in its current state. |
| Pull Request Summary | /gemini summary |
Provides a summary of the current pull request in its current state. |
| Comment | @gemini-code-assist | Responds in comments when explicitly tagged, both in issue comments and review comments. |
| Help | /gemini help |
Displays a list of available commands. |
Customization
To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.
Limitations & Feedback
Gemini Code Assist is currently in preview and may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments to provide feedback.
You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.
Footnotes
-
Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩
There was a problem hiding this comment.
Code Review
This pull request introduces a new text chunking utility for Text-to-Speech (TTS) to improve streaming latency. The core logic is in a new chunkTTSInput generator function, and it's integrated into the message processing queue. A new playground component is also added for testing. There are some critical issues in the new logic that need to be addressed.
packages/stage-ui/src/components/Scenarios/Providers/SpeechStreamingPlayground.vue
Outdated
Show resolved
Hide resolved
packages/stage-ui/src/components/Scenarios/Providers/SpeechStreamingPlayground.vue
Outdated
Show resolved
Hide resolved
packages/stage-ui/src/components/Scenarios/Providers/SpeechStreamingPlayground.vue
Outdated
Show resolved
Hide resolved
Co-authored-by: Neko <neko@ayaka.moe>
* feat(stage-ui): better tts input chunking * chore: remove unused audio analyser * fix: words count * fix: use ctx.data instead --------- Co-authored-by: Neko <neko@ayaka.moe>
* feat(stage-ui): better tts input chunking * chore: remove unused audio analyser * fix: words count * fix: use ctx.data instead --------- Co-authored-by: Neko <neko@ayaka.moe>
Description
chunkTTSInputinpackages/stage-ui/src/utils/tts.tsto pre-process text into chunks before streaming them to the speech provider, to reduce the first-chunk wait time while preferring to generate speech with complete sentences. (Please allow me to book a slot in our DevLog to explain this.)