fix: config template updated for streaming service as deepgram

[01PrathamS]

Updating config template with streaming support as deepgram (as parakeet doesn't support streaming)

Addressing #284

Summary by CodeRabbit

Release Notes

• New Features

  • Memory extraction feature added to automatically identify and structure key information from conversations.

• Chores

  • Speech-to-text model configuration enhanced with upgraded provider support and improved streaming transcription parameters including smart formatting and punctuation.

[coderabbitai]

📝 Walkthrough

Walkthrough

The changes update a speech-to-text plugin script with explicit UTF-8 encoding for file operations, and replace the Parakeet streaming transcription model configuration with Deepgram Nova 3 streaming, including new API credentials, restructured operations, and updated result field mappings.

Changes

Cohort / File(s)	Summary
Plugin Script Encoding backends/advanced/scripts/create_plugin.py	Adds explicit encoding="utf-8" parameter to three write_text() calls for __init__.py, plugin.py, and README.md file generation.
Speech-to-Text Model Configuration config/config.yml.template	Replaces stt-parakeet-stream model with stt-deepgram-stream; updates provider from Parakeet to Deepgram with new WebSocket endpoint and API key credential. Restructures streaming operations to use single query block with Deepgram parameters (model, language, formatting, encoding, sample_rate, channels). Updates end-of-stream signal from stop to CloseStream. Remaps result extraction fields to use channel.alternatives[0] with new data paths for transcript, words, and paragraphs. Adds new memory.extraction block with structured fact extraction prompt.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~25 minutes

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately describes the main change: updating the config template to use Deepgram for streaming service, which is the primary objective of the PR.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.

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

✨ Finishing touches

• 📝 Generate docstrings

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

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

[AnkushMalaker]

@coderabbitai review

[coderabbitai]

✅ Actions performed

Review triggered.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.

[AnkushMalaker]

LGTM! Thanks!