Commit tts demo audio clips for review without local playback

Reference and cloned-output wav files are small (~2MB each) and are the
actual evidence of the voice-clone+translate demo working, so track them
despite the general *.wav exclusion.

Co-Authored-By: Claude Sonnet 5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01Y1oQ4h4WE4R9KRvMQLr5Ey

.gitignore

@@ -29,3 +29,9 @@ outputs/
 # Full-size source video — keep locally, commit only compressed derivatives
 media-files/video/*.mp4
 !media-files/video/*-480p.mp4
+
+# Small demo evidence clips — keep committed so results are reviewable
+# without local playback (unlike demucs's separated/ stems, which stay local)
+!tts/reference/*.wav
+!tts/output/
+!tts/output/*.wav