diff --git a/CHANGELOG.md b/CHANGELOG.md
index 1e09dfb..b371c64 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -100,10 +100,10 @@
 - **Edit emotions: every clip in the list, and the line follows the clip (编辑情绪：下拉列出全部音频，换音频台词跟着换)**: a box that already held a file listed only the files whose names contain it (the browser filters by the box's text), so a character's other clips seemed missing. Pressing the box now lists all of them, with the current file shown greyed; leaving without a pick keeps it. Picking another clip no longer keeps the old clip's line beside it: the line the node already reads for that clip is filled in, otherwise the line the node finds for the clip (its .txt / .lab or the annotation list) is filled in, with Anomalous_TTS from interface 14 on (older versions leave the box blank and the node reads the line when it speaks); going back to the first clip brings its line back. A blank line box shows, greyed, the line the node uses for its clip.
 - **Search a character's clips by line (按台词找参考音频)**: Edit emotions has a search box above the clips: type part of a line or a file name and the matches list with their lines, to hear (▶), **Use as main voice** or **Add as emotion** (a new emotion row with that clip and its line, ready for its name). Searching lines needs Anomalous_TTS interface 14; with an older version it searches file names.
 
-## v1.59.1 Beta (GPT-SoVITS v3 / v4 voices) — unreleased
+## v1.59.1 Beta (GPT-SoVITS v3 / v4 voices) — 2026-10-17
 
 ### 🎙️ Character voices: v3 and v4 models (角色语音支持 v3 / v4 模型)
-- **Anomalous_TTS now runs GPT-SoVITS v3 and v4 models** (update it in ComfyUI Manager), and this version of the plugin goes with it: v3 / v4 files are imported instead of refused, and a file whose version cannot be read is refused with a clearer reason. With an Anomalous_TTS too old for v3 / v4, the import says to update it.
+- **Anomalous_TTS now runs GPT-SoVITS v3 and v4 models** (its first stable release, [v1.0.0](https://github.com/DemonGatanjieu/Anomalous_TTS/releases/tag/v1.0.0); update it in ComfyUI Manager), and this version of the plugin goes with it: v3 / v4 files are imported instead of refused, and a file whose version cannot be read is refused with a clearer reason. With an Anomalous_TTS too old for v3 / v4, the import says to update it.
 - **Each character card shows its GPT-SoVITS model version** (v2, v2ProPlus, v4…) after the name (角色卡片显示模型版本).
 - **v3 and v4 need their own official base model and vocoder** (v4 about 0.8 GB, v3 about 1.2 GB). **GPT-SoVITS Settings** lists them, with their own labels, only when one of your characters uses that version, and the import card says what is still missing for the model you dropped in. They download on first use too; a GPT-SoVITS package already on the computer is used as it is.