OpenAI speech detect

[dariopellegrino00]

No description provided.

[Copilot]

Pull Request Overview

This PR implements OpenAI speech detection functionality to track when OpenAI starts and stops speaking during audio streaming sessions. The implementation adds event-driven notifications for speech state changes to enable better handling of conversation flow and interruptions.

• Adds speech state tracking with m_openai_speaking and m_response_audio_done flags
• Implements speech start/stop detection logic in the audio playback pipeline
• Introduces new event types for notifying external systems of speech state changes

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.

File	Description
openai_audio_streamer_glue.cpp	Core speech detection logic, state tracking, and event notification methods
mod_openai_audio_stream.h	Event constant definitions for speech start/stop notifications
mod_openai_audio_stream.c	Event registration and cleanup for new speech detection events

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[Copilot]

Setting inuse = 0 after clearing the buffer could cause incorrect behavior. The inuse variable represents the actual bytes in the buffer, and after switch_buffer_zero() it should reflect the buffer's empty state, but this assignment might interfere with the subsequent logic that depends on the actual buffer size.

Suggested change

	inuse = 0;
	inuse = switch_buffer_inuse(tech_pvt->playback_buffer);

[Copilot]

[nitpick] The condition combines two state checks that could be confusing. Consider adding a comment explaining when both conditions are true simultaneously, or create a helper method like should_stop_speech() to encapsulate this logic.