Inworld tts auto mode

[ianbbqzy]

auto_mode to be added to config param in a separate PR when word tokenizer and user-controlled manual flushes are supported. For now, auto_mode should enhance quality and naturalness of agent response

Summary by CodeRabbit

• Updates
  • Auto Mode enabled for Inworld TTS operations.
  • Streaming text now flushes context more frequently during transmission, improving responsiveness and reducing latency for long messages.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

---

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.

---

Ian Lee seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[coderabbitai]

📝 Walkthrough

Walkthrough

The Inworld TTS plugin now sets autoMode to True when creating a WebSocket context and calls flush_context inside the per-chunk sending loop (after each ~1000-character chunk) instead of only once after all chunks are sent.

Changes

Cohort / File(s)	Summary
Inworld TTS Auto Mode & Flushing livekit-plugins/livekit-plugins-inworld/livekit/plugins/inworld/tts.py	Set pkt["create"]["autoMode"] = True when creating the WebSocket context. Moved flush_context into the per-chunk send loop so it runs after each ~1000-character chunk; marked as temporary until auto_mode is released.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Poem

I nibble bytes and polish code with care,
AutoMode set true — a bright new hare!
Flush after chunks, hop by hop I bound,
Temporary steps where changes are found,
A rabbit cheers at fixes neatly sound. 🐇✨

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 33.33% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title '[Draft] inworld tts auto mode' clearly describes the main change: enabling auto mode for inworld TTS. It directly relates to the implementation of auto_mode behavior in the TTS module.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

• 📝 Generate docstrings

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@livekit-plugins/livekit-plugins-inworld/livekit/plugins/inworld/tts.py`:
- Around line 358-359: The unconditional setting of pkt["create"]["autoMode"] =
True (in the TTS packet creation where SentenceTokenizer is used) should be
removed or gated to avoid sending an undocumented field; change this to follow
the existing pattern used for timestampType/applyTextNormalization (only set
autoMode if the corresponding config value is provided/is_given()), or otherwise
remove the line and/or wrap the send in graceful error handling to ignore server
rejections related to unknown fields; update the code around the packet
construction in the TTS send/create logic (the block that references
SentenceTokenizer and pkt["create"]) to implement the conditional check or
removal.

🧹 Nitpick comments (1)

livekit-plugins/livekit-plugins-inworld/livekit/plugins/inworld/tts.py (1)

1081-1086: Gate flush_context behind a flag when autoMode is enabled.

Currently, flush_context is called after every sentence token. With autoMode enabled (which is unconditional), the Inworld server handles buffering internally, making per-token flushes redundant and increasing message overhead. Consider gating this behind a flag or removing it entirely in favor of explicit flushes via user input or end_input().

📜 Review details

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 618c4b530c21160498255af0e3a0946c3dd24f04 and 05953141b98c71fab26a65a62ef7b4f6a71fe57d.

📒 Files selected for processing (1)

• livekit-plugins/livekit-plugins-inworld/livekit/plugins/inworld/tts.py

🧰 Additional context used

📓 Path-based instructions (1)

**/*.py

📄 CodeRabbit inference engine (AGENTS.md)

**/*.py: Format code with ruff
Run ruff linter and auto-fix issues
Run mypy type checker in strict mode
Maintain line length of 100 characters maximum
Ensure Python 3.9+ compatibility
Use Google-style docstrings

Files:

• livekit-plugins/livekit-plugins-inworld/livekit/plugins/inworld/tts.py

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)

• GitHub Check: unit-tests

_{✏️ Tip: You can disable this entire section by setting review_details to false in your review settings.}

[devin-ai-integration]

✅ Devin Review: No Issues Found

Devin Review analyzed this PR and found no potential bugs to report.

View in Devin Review to see 5 additional flags.

[toubatbrian]

cc @theomonnom @longcw

[tinalenguyen]

thanks for the PR! are you still unable to sign the CLA?

[ianbbqzy]

thanks for the PR! are you still unable to sign the CLA?

Doesn't look like so. This comment from CLAassistant also seems weird

Ian Lee seems not to be a GitHub user.

[tinalenguyen]

I used to see those when pushing from different devices and I needed to re-authorize my GH account for one, not sure if that's the case here