The science of speech. The character of a voice.

Meet the Simba speech synthesis models. Expressive English with Simba 3.2, multilingual synthesis with Simba 3.0. Hear the difference, then build with it.

Read the Simba 3.2 launch and migration guide

Shared foundation

Streaming-native
Self-serve voice cloning
24kHz output

Speech synthesis modelRecommended

Simba 3.2

model="simba-3.2"

Streaming-native English speech with rich expressivity

Our flagship streaming-native English model. Audio streams as it generates, with finer-grained emotional control, SSML prosody, and every English voice in the catalog, your cloned voices included.

Read the Simba 3.2 announcement

Delivery: Streaming-native
Languages: English
Voice cloning: Self-serve
Sample rate: 24kHz

Capabilities

Streaming-native architecture
Emotional expression
Every English voice in the catalog
Self-serve voice cloning
SSML prosody control

Expression study

Emotion Control

Direct the same line toward neutral, calm, cheerful, energetic, or sad delivery. Simba 3.2 shapes rhythm and tone with SSML emotion control.

Simba 3.2 first-byte latency on our production US East streaming path measured 56 ms p50 and 102 ms p90 on 15 Sep 2026. This is not first audible sound or a latency guarantee. Streaming API details ↗

Same text · five directionssimba-3.2

“The quick brown fox jumped over the lazy dog.”

Neutral01

Calm02

Cheerful03

Energetic04

Sad05

Speech synthesis model

Simba 3.0

model="simba-3.0"

Streaming-native synthesis beyond English

Streaming-native synthesis in English, German, Spanish, French, Italian, and Brazilian Portuguese. Routing happens automatically by language, and zero-shot voice cloning works self-serve.

Delivery: Streaming-native
Locales: 7
Voice cloning: Self-serve
Sample rate: 24kHz

Capabilities

Streaming-native architecture
7 supported locales
Zero-shot voice cloning
Emotional expression
SSML prosody control

Language range

Multilingual Synthesis

Use Simba 3.0 for English, German, Mexican Spanish, French, Italian, and Brazilian Portuguese. Each sample uses a voice cataloged for that locale.

Need a language outside the self-serve set? We cover 48 languages in total - talk to us about reaching them.

Locale decksimba-3.0

languages · 7 locales

en
EnglishPrepared catalog sample
de-DE
GermanPrepared catalog sample
es-MX
Spanish (Mexico)Prepared catalog sample
es-ES
Spanish (Spain)Supported; no preview sample
fr-FR
FrenchPrepared catalog sample
it-IT
ItalianPrepared catalog sample
pt-BR
Portuguese (BR)Prepared catalog sample

6 prepared catalog samples. Both Spanish locales support catalog voices; check voice compatibility before synthesis.Check GET /v1/voices

Voice identity

Zero-Shot Voice Cloning

Create a reusable voice from a short, consented reference clip. Self-serve cloning on Simba 3.2 and Simba 3.0 preserves the speaker's identity across new scripts.

Explore voice cloning

Reference and synthesissimba-3.0 · zero-shot

01 / Reference

Original speaker

Ready to listen0:00

02 / Clone

Simba 3.0 output

Ready to listen0:00

Two different recordings. Only clone a voice you have permission to use.

One synthesis surface

All models, one API

Access every model through the same endpoint. Choose the model and a voice compatible with its language support.

simba-3.2: Expressive English and finer emotional direction
simba-3.0: Multilingual synthesis across the self-serve locale set

Text-to-Speech API Documentation

python

from speechify import Speechify

client = Speechify()  # uses SPEECHIFY_API_KEY env var

response = client.tts.audio.speech(
    input='<speak><speechify:style emotion="cheerful">Every moment of light and dark is a miracle.</speechify:style></speak>',
    voice_id="geffen_32",
    model="simba-3.2",
    audio_format="mp3",
)

with open("output.mp3", "wb") as f:
    f.write(response.audio_data)

Built on speech. Open to possibility.

Make something
worth listening to.

Start building Talk to sales