Python 3.13 misaki installation workaround

README.md

@@ -1,5 +1,12 @@
 # Kitten TTS 😻
 
+THIS IS A MODIFIED VERSION
+- modified to be able to install on Python 3.13
+- modified to remove a dependency on Torch
+
+It was modified in order to explore using it in: 
+https://github.com/createcandle/voco
+
 <img width="607" height="255" alt="Screenshot 2026-02-18 at 8 33 04 PM" src="https://github.com/user-attachments/assets/f4646722-ba78-4b25-8a65-81bacee0d4f6" />
 
 

kittentts/get_model.py

@@ -34,7 +34,7 @@ def generate(self, text, voice="expr-voice-5-m", speed=1.0, clean_text=False):
         Returns:
             Audio data as numpy array
         """
-        print(f"Generating audio for text: {text}")
+        #print(f"Generating audio for text: {text}")
         return self.model.generate(text, voice=voice, speed=speed, clean_text=clean_text)
 
     def generate_to_file(self, text, output_path, voice="expr-voice-5-m", speed=1.0, sample_rate=24000):

pyproject.toml

@@ -19,13 +19,13 @@ classifiers = [
 ]
 dependencies = [
     "num2words",
-    "spacy",
     "espeakng_loader",
-    "misaki[en]>=0.9.4",
     "onnxruntime",
     "soundfile",
     "numpy",
     "huggingface_hub",
+    "transformers",
+    "phonemizer-fork"
 ]
 
 [project.urls]

requirements.txt

@@ -1,8 +1,8 @@
 num2words
-spacy
 espeakng_loader
-misaki[en]>=0.9.4
 onnxruntime
 soundfile
 numpy
 huggingface_hub
+transformers
+phonemizer-fork

setup.py

@@ -24,19 +24,22 @@
         "Programming Language :: Python :: 3.10",
         "Programming Language :: Python :: 3.11",
         "Programming Language :: Python :: 3.12",
+        "Programming Language :: Python :: 3.13",
         "Topic :: Multimedia :: Sound/Audio :: Speech",
         "Topic :: Scientific/Engineering :: Artificial Intelligence",
     ],
     python_requires=">=3.8",
     install_requires=[
         "num2words",
-        "spacy",
+        #"spacy",
         "espeakng_loader",
-        "misaki[en]>=0.9.4",
+        #"misaki@git+https://github.com/createcandle/misaki",
         "onnxruntime",
         "soundfile",
         "numpy",
         "huggingface_hub",
+        "transformers",
+        "phonemizer-fork",
     ],
     keywords="text-to-speech, tts, speech-synthesis, neural-networks, onnx",
     project_urls={