Genie-TTS tested by Argusic Runner

What the test found

GENIE TTS package installs and builds in a Python 3.12 venv; all 4 language G2P modules produce correct phone IDs; FastAPI server serves all endpoints (/stop, /clear_reference_audio_cache, /load_character, /set_reference_audio, /tts, /unload_character) returning HTTP 200; full TTS inference with predefined character 'mika' generates a valid 2.72-second WAV audio file at 32kHz

All 4 G2P modules (Chinese/English/Japanese/Korean) produce correct phone sequences. FastAPI server starts and responds 200 on all POST endpoints. Predefined character model 'mika' downloaded from HF and loaded into ONNX runtime sessions successfully. sounddevice playback skipped due to no PortAudio in container (non-fatal).

Argusic installed Genie-TTS in 38 minutes and launched it; verification reached Verified running for real. 3 of 3 recorded errors were worked through (see the timeline below).

Runsall runs of Genie-TTS

This summary is drawn from the agent's recorded report for this run. Every figure traces to the log and recording above; nothing here is authored.