Oct 23 19 days ago 21:24 Card: point to v2 (the -v2 name now belongs to the new repo) Oct 12 30 days ago 08:13 Space voice order: best to weakest (male_voxcpm_2, female_voxcpm_2, male_voxcpm_1, female_voxcpm_1); default male_voxcpm_2 07:48 Defaults for every voice: temperature 0.8, tempo 0.8 (voices_index, voices.json, card); samples of the new voices regenerated at these defaults 07:18 female_voxcpm_1: 90 Hz high-pass post-filter (removes p pops learned from its VoxCPM2 reference clip): highpass_hz in voices_index/voices.json, note in the card, filtered samples Oct 11 19:26 ATTRIBUTION: base model section in English under its new name; samples table: labels and relative links fixed 19:11 Renamed from Matcha-TTS-PL-v2 to Matcha-TTS-PL; base model is Matcha-TTS-PL-Pretrain Oct 10 04:34 Three more synthetic voices (female_voxcpm_1, male_voxcpm_2, female_voxcpm_2): one Matcha model + own HiFi-GAN each, onnx/<voice>/, voices_index.json; card and attribution for four voices Oct 5 20:23 Card: English detection convention and switch 20:23 English detection: phrase matching, Polish words (SGJP via Morfeusz 2) read as Polish unless listed or in braces 10:29 style_map: neutral = 8 (mid-range 路 normal); 7 is mid-range 路 slow 路 question 10:23 Questions: keep '?' glued (as in training); neutral style 8 10:09 Default temperature 1.0 10:04 Browser front end: current cleaner.js / english.js 08:00 Voice renamed fav2 -> male_voxcpm_1; question-mark tip; browser cleaner keeps 'word ?' Oct 2 18:40 Corrected KPWr credits; link the training dataset 14:45 Matcha-TTS-PL v2: fav2 voice distilled from VoxCPM2, English words, fav2 vocoder 14:43 initial commit 
