From 79d149f3b8754b10101e6011d3e2508c6f86ca2a Mon Sep 17 00:00:00 2001
From: Kingston-SCYD <odscheidel@gmail.com>
Date: Tue, 9 Jun 2026 16:06:54 -0500
Subject: [PATCH] feat: add minimax mode to /api/mc/tts endpoint

---
 ttstoy/webui/app.py | 33 ++++++++++++++++++++++++++++++++-
 1 file changed, 32 insertions(+), 1 deletion(-)

diff --git a/ttstoy/webui/app.py b/ttstoy/webui/app.py
index b98c372..33a2099 100644
--- a/ttstoy/webui/app.py
+++ b/ttstoy/webui/app.py
@@ -686,7 +686,7 @@ def api_mc_tts():
         r"(<a?:\w+:\d+>|:[A-Za-z0-9_]+:|[\U0001F300-\U0001FAFF\U00002600-\U000026FF\U00002700-\U000027BF][\uFE0E\uFE0F]?(?:\u200D[\U0001F300-\U0001FAFF\U00002600-\U000026FF\U00002700-\U000027BF][\uFE0E\uFE0F]?)*)"
     )
     VOICE_SWITCH_RE = re.compile(r"\[([a-zA-Z]+)(?:\|([^\]]+))?\]")
-    VALID_MODES = {"chatterbox", "clone", "dectalk", "morshu", "vox"}
+    VALID_MODES = {"chatterbox", "clone", "dectalk", "morshu", "vox", "minimax"}
     # Turbo tokens — these look like voice switches but should pass through to chatterbox
     TURBO_TOKENS = {"laugh", "chuckle", "sigh", "gasp", "cough", "cry", "groan", "yawn", "sniff"}
 
@@ -738,6 +738,37 @@ def api_mc_tts():
             r = requests.get("http://127.0.0.1:33003/say", params={"text": seg_text, "pack": pack}, timeout=60)
             r.raise_for_status()
             return r.content
+        elif mode == "minimax":
+            api_key = gcfg.get("minimax_api_key")
+            if not api_key:
+                raise RuntimeError("MiniMax API key not configured in ttstoy")
+            model = gcfg.get("minimax_model", "speech-01-turbo")
+            payload = {
+                "model": model,
+                "text": seg_text,
+                "stream": False,
+                "language_boost": "auto",
+                "output_format": "hex",
+                "voice_setting": {
+                    "voice_id": voice or "male-qn-qingse",
+                    "speed": 1,
+                    "vol": 1,
+                    "pitch": 0,
+                },
+                "audio_setting": {
+                    "sample_rate": 32000,
+                    "bitrate": 128000,
+                    "format": "mp3",
+                },
+            }
+            headers = {"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"}
+            r = requests.post("https://api.minimax.io/v1/t2a_v2", json=payload, headers=headers, timeout=60)
+            r.raise_for_status()
+            resp = r.json()
+            audio_hex = resp.get("data", {}).get("audio", "")
+            if not audio_hex:
+                raise RuntimeError("MiniMax returned no audio")
+            return bytes.fromhex(audio_hex)
         else:
             raise RuntimeError(f"Unknown mode: {mode}")