---

license: apache-2.0 library_name: onnx pipeline_tag: text-to-speech language: [en, de, es, it, pt, ru, hi, da, fi, fr, ja, ko, nl, sv, sw, tr] tags: [text-to-speech, onnx, int8, safetensors, pytorch, kittentts]

Kitten TTS 2 Mini

82M parameters, 16 languages, 118 voices, 24 kHz. ONNX with int8 weights (82.7 MB; runs on CPU, no PyTorch needed), and the float32 PyTorch weights.

Usage

pip install kittenml

from kittenml import KittenTTS
import soundfile as sf

m = KittenTTS("KittenML/kitten-tts-2-mini")          # or the path of this directory
print(m.available_voices)                 # {"en": ["Eleanor", ...], "de": ["Lena", ...], ...}
for v in m.voices("de"):                  # name and description of each voice
    print(v["name"], v["description"])

audio = m.generate("Hello there, this is Kitten TTS 2 Mini.", voice="Eleanor")
audio = m.generate("Guten Morgen! Wie geht es dir?", voice="Felix", speed=1.1)    # the voice sets the language
audio = m.generate("Guten Morgen! Wie geht es dir?", language="de")          # the language's first voice (Lena)
sf.write("out.wav", audio, 24000)

m = KittenTTS("KittenML/kitten-tts-2-mini", backend="torch")                  # PyTorch, float32 weights
m = KittenTTS("KittenML/kitten-tts-2-mini", backend="torch", device="cuda")   # on a GPU
audio = m.generate("Hello there, this is Kitten TTS 2 Mini.", voice="Eleanor")

backend="onnx" (the default) runs the int8 ONNX graph with ONNX Runtime; backend="torch" runs the float32 weights in PyTorch. Only the files the selected backend needs are downloaded.

voice takes a name from the tables below (any case: "Felix" or "felix") or its description. Without voice, a language's first voice speaks. In this model the English voices also speak German, Spanish, Italian, Portuguese and Russian under the same name: pass language= to choose (m.generate("Hallo.", voice="Eleanor", language="de")); without it, English.

Voices

Each voice is a first name; its voice pack is stored under the lowercase name in voices_<lang>.npz. The English voices carry the names of the same characters in KittenTTS 2.

English (en)

voice	description
Eleanor	Somber historian (female)
Martha	Weathered radio journalist (female)
Serena	Gentle yoga instructor (female)
Iris	Hushed museum guide (female)
Grace	Soothing nurse (female)
Edith	Elderly grandmother soft (female)
Diana	Stern military officer (female)
Otis	Rumbling blues musician (male)
Vincent	Hushed museum guide (male)
Max	Energetic sports announcer (male)
Danny	Cheerful podcast host (male)

German (de)

voice	description
Lena	Female 1
Greta	Female 2
Jonas	Male 1
Felix	Male 2
Eleanor	Somber historian (female)
Martha	Weathered radio journalist (female)
Serena	Gentle yoga instructor (female)
Iris	Hushed museum guide (female)
Grace	Soothing nurse (female)
Edith	Elderly grandmother soft (female)
Diana	Stern military officer (female)
Otis	Rumbling blues musician (male)
Vincent	Hushed museum guide (male)

Spanish (es)

voice	description
Lucia	Female 1
Carmen	Female 2
Javier	Male 1
Diego	Male 2
Eleanor	Somber historian (female)
Martha	Weathered radio journalist (female)
Serena	Gentle yoga instructor (female)
Iris	Hushed museum guide (female)
Grace	Soothing nurse (female)
Edith	Elderly grandmother soft (female)
Diana	Stern military officer (female)
Otis	Rumbling blues musician (male)
Vincent	Hushed museum guide (male)

Italian (it)

voice	description
Giulia	Female 1
Chiara	Female 2
Marco	Male 1
Luca	Male 2
Eleanor	Somber historian (female)
Martha	Weathered radio journalist (female)
Serena	Gentle yoga instructor (female)
Iris	Hushed museum guide (female)
Grace	Soothing nurse (female)
Edith	Elderly grandmother soft (female)
Diana	Stern military officer (female)
Otis	Rumbling blues musician (male)
Vincent	Hushed museum guide (male)

Portuguese (Brazil) (pt)

voice	description
Beatriz	Female 1
Camila	Female 2
Rafael	Male 1
Thiago	Male 2
Eleanor	Somber historian (female)
Martha	Weathered radio journalist (female)
Serena	Gentle yoga instructor (female)
Iris	Hushed museum guide (female)
Grace	Soothing nurse (female)
Edith	Elderly grandmother soft (female)
Diana	Stern military officer (female)
Otis	Rumbling blues musician (male)
Vincent	Hushed museum guide (male)

Russian (ru)

voice	description
Anastasia	Female 1
Natasha	Female 2
Dmitri	Male 1
Ivan	Male 2
Eleanor	Somber historian (female)
Martha	Weathered radio journalist (female)
Serena	Gentle yoga instructor (female)
Iris	Hushed museum guide (female)
Grace	Soothing nurse (female)
Edith	Elderly grandmother soft (female)
Diana	Stern military officer (female)
Otis	Rumbling blues musician (male)
Vincent	Hushed museum guide (male)

Hindi (hi)

voice	description
Priya	Female 1
Ananya	Female 2
Arjun	Male 1
Vikram	Male 2

Danish (da)

voice	description
Freja	Female 1
Mads	Male 1
Rasmus	Male 2

Finnish (fi)

voice	description
Aino	Female 1
Helmi	Female 2
Eero	Male 1
Juho	Male 2
Ville	Male 3

French (fr)

voice	description
Camille	Female 1
Chloe	Female 2
Manon	Female 3
Louis	Male 1
Antoine	Male 2
Julien	Male 3
Mathis	Male 4
Etienne	Male 5

Japanese (ja)

voice	description
Yuki	Female 1
Sakura	Female 2
Haruka	Female 3
Aoi	Female 4
Haruto	Male 1

Korean (ko)

voice	description
Seoyeon	Female 1
Minjun	Male 1
Jihoon	Male 2

Dutch (nl)

voice	description
Sanne	Female 1
Fenna	Female 2
Daan	Male 1
Bram	Male 2
Joris	Male 3

Swedish (sv)

voice	description
Astrid	Female 1
Elsa	Female 2
Oskar	Male 1

Swahili (sw)

voice	description
Zawadi	Female 1
Neema	Female 2
Baraka	Male 1
Juma	Male 2

Turkish (tr)

voice	description
Elif	Female 1
Zeynep	Female 2

License

Apache-2.0 (the models, voice packs and the bundled G2P files).