Add Qwen 3 TTS support for Simplismart-livekit plugin

[simplipratik]

Summary

This PR adds Qwen 3 TTS support to the SimpliSmart LiveKit plugin while maintaining full backwards compatibility with the existing Orpheus TTS implementation.

Changes

• Added qwen-tts to TTSModels in models.py
• Created QwenTTSOptions class for Qwen 3 specific configuration (language, leading_silence)
• Updated TTS class to accept an options parameter that can be either SimplismartTTSOptions or QwenTTSOptions
• Updated ChunkedStream._run() to use type-based payload construction for the appropriate API format
• Added comprehensive error handling and logging (consistent with stt.py patterns)
• Exported QwenTTSOptions and SimplismartTTSOptions from the module

Usage

Legacy Orpheus (backwards compatible)

from livekit.plugins.simplismart import TTS

tts = TTS()  # Uses default Orpheus model
tts = TTS(model="canopylabs/orpheus-3b-0.1-ft", voice="tara")

Qwen 3 TTS

from livekit.plugins.simplismart import TTS, QwenTTSOptions

tts = TTS(
    model="qwen-tts",
    voice="Chelsie",
    base_url="https://api.simplismart.live/v1/audio/speech",
    options=QwenTTSOptions(language="English", leading_silence=True)
)

Files Changed

• livekit-plugins/livekit-plugins-simplismart/livekit/plugins/simplismart/__init__.py
• livekit-plugins/livekit-plugins-simplismart/livekit/plugins/simplismart/models.py
• livekit-plugins/livekit-plugins-simplismart/livekit/plugins/simplismart/tts.py

[u9g]

Getting rid of these four will break existing users, it's a breaking change.

[u9g]

Setting model="qwen-tts" or base_url=QWEN_BASE_URL without options=QwenTTSOptions(...) sends Orpheus-format JSON to the Qwen endpoint. Is there a better way we can approach the is_qwen check?

[u9g]

sample_rate is configurable, but is sent to the server in neither Qwen nor Orpheus mode, is this intentional?

[u9g]

this and the other 6 logger.info's should be logger.debug's, otherwise this will be very noisy, and out of line with the way the other plugins log.