Àwọn àwòrán àkọlé ìṣàmúlò-ètò

Gbogbo àwọn módèlè̀ TTS nínú àwọn pálẹ́ẹ̀lì wa ní afẹ́fẹ́ pẹ̀lú́ àwọn lísínsì tí a fẹ́. MIT, Apache 2.0 — kò ní ìjádé, kò ní ìgbàyélú, kò ní owó lísínsì tí a kò lè fi hàn. Lò nípá API wà, tàbí lójútó wọn nípá ińfráẹ́ẹ̀kì rẹ̀.

Àwọn Ìṣàmúlò-ètò Mìíràn Àwọn Àwọn Àwọn Àwọn Apache Àwọn Ààyè-iṣẹ́ GitHub

Àtòjọ-ètò TTS Fẹ́ẹ̀lì Àwọn Àkọsílẹ̀ API

_Yanju

0/500

Free pẹlu Kokoro, Piper, VITS, MeloTTS

Àwọn àwòrán tí o ti ṣẹ̀dà tí o bá han níbẹ̀

Ṣí àwọn àyọkà ìṣàmúlò-ètò TTS kíki

Àwọn Ìfànẹ́ Ìṣàmúlò-ètò TTS

Kini idi ti àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn

Gbogbo àwọn ààyè-iṣẹ́ àìfilọ́lẹ̀

Every model on TTS.ai uses a permissive open-source license. No proprietary black boxes, no vendor lock-in, no unexpected licensing fees.

Apache

Àwọn àwọn àwòrán ní ìṣẹ̀dà láti inú MIT tàbí Apache 2.0, àwọn ìṣẹ̀dà àìṣiró-ìṣàfilọ́lẹ̀ tí o jú. Lò nípa ìṣowo, ìṣàfarawe, ìṣàfilọ́lẹ̀ padà— kò ní àwọn ìgbàjójú.

Àwọn Ààyè-iṣẹ́

Ṣàfihàn àwọn módè́lì àti ìṣàmúlò-ètò fún ọ̀rọ̀rọ̀ rẹ̀. Ìdáràn fún àwọn ààyè-iṣẹ́ rẹ̀, àwọn ìṣàfihàn, àti àwọn àgbègbè. Kò ní ìgbà tí a fẹ́.

GPU tí a fi pamọ́

Àwọn móòdù ní pàtó fún NVIDIA GPÙ̀ àti ìrànwọ́ CUDA. Piper̀ láàmúlò nípa CPÙ lọ́wọ́lọ́wọ́. Móòdù àwọn púpọ̀ ní niló 2-8GB VRAM fún ìṣàfarawé.

Àwọn Àwọn Àwọn Àwọn Àwọn

Active open-source communities maintain and improve these models. Contributions welcome — submit bugs, improvements, and new voices on GitHub.

Àwọn Ìṣàmúlò-ètò

Gbogbo àwọn ìṣàmúlò-ètò náà náà gba ìlò-ètò fún iṣẹ́ nínú àwọn lísínsì wọn. Ṣẹ̀dà àwọn nǹkan, tá àwọn iṣẹ́, àti ìṣàmúlò-ètò fún iṣẹ́ tí kò ní ẹ̀yàn àwọn ẹ̀yàn ìlò-ètò.

Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn

Àwọn ìṣàmúlò-ètò, àwọn ìṣẹ́dá rẹ̀, àti ohun tí o ṣè

Kokoro

Free

Lightweight 82M parameter model delivering studio-quality speech with blazing-fast inference.

Fast 5/5

Ó dara jù fún: Apache 2.0 — ìṣàmúlò-ètò ọ̀fẹ̀ tí o dara jú, 82M àwọn ìṣàmúlò-ètò, rọ́ọ̀nù fún alábòójútó

Àwọn ààyè-iṣẹ́ Kokoro

Piper

Free

A fast, local neural text to speech system optimized for Raspberry Pi and embedded devices.

Fast 3/5

Ó dara jù fún: MIT — CPU-kan, to dara fún àwọn àpá-irinṣẹ́ ìsàlẹ̀-ilà àti àwọn ààyè-iṣẹ́ ìṣàfarawe-ẹni-ní

Àwọn ààyè-iṣẹ́ Piper

VITS

Free

Conditional variational autoencoder with adversarial learning for end-to-end text-to-speech.

Fast 3/5

Ó dara jù fún: MIT — ààyè ìdájọ́ tí a lò láti mú àwọn móòdù ìsàlẹ̀-ilà láti sà

Àwọn ààyè-iṣẹ́ VITS

Bark

Standard

Transformer-based text-to-audio model that generates realistic speech, music, and sound effects.

Slow 4/5

Ó dara jù fún: MIT — àwọn ìṣàfilọ́lẹ̀ ìṣàfilọ́lẹ̀ àwòrán tí a kò lè gbọ́ nínú TTS ìṣàfarawé

Àwọn ààyè-iṣẹ́ Bark

Tortoise TTS

Premium

Multi-voice text-to-speech focused on quality with autoregressive architecture.

Slow 5/5 Àwọn Àmì-ìwé

Ó dara jù fún: Apache 2.0 — ìṣàmúlò-ètò ìṣàfihàn ìṣàfihàn tí a kọ̀ nípa ìṣàfihàn

Àwọn ààyè-iṣẹ́ Tortoise TTS

OpenVoice

Premium

Instant voice cloning with granular control over style, emotion, and accent.

Medium 4/5 Àwọn Àmì-ìwé

Ó dara jù fún: MIT — ìṣàfarawé àwọn ìròyìn àwọn ìsàlẹ̀-ilà àti ìpéwọ̀n ìṣàfarawé àwọn ìṣàmúlò-ilà

Àwọn ààyè-iṣẹ́ OpenVoice

Bii o ṣe le lo TTS orisun-aṣiṣii

Ló àwọn API ààyè-iṣẹ́ wà tàbí láti ló àwọn àwọn ìṣàmúlò-ètò rẹ̀

Wá Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn

Browse our catalog of 20+ open-source TTS models. Each model page shows the license, architecture, capabilities, and self-hosting requirements.

Wá nínú Awáròyìn Rẹ̀

Ṣayẹwo àwọn módè́ẹ̀lì fún TTS.ai nípa ìṣàfihàn ohunkohun. Àwọn sáà GPU wa nípa ìṣàfarawé ìṣàfarawé láti jẹ́ pé o lè ṣàyẹwo ìgúnrégé nípa ìṣàfihàn ara-ẹni.

Àwọn Ìṣàfilọ́lẹ̀ Àwọn Àwọn Ìṣàfilọ́lẹ̀

Clone model repos from GitHub and run locally, or use our hosted API for production. Self-hosting gives full control; our API provides managed infrastructure.

Ṣẹ̀dá Àtòjọ-ètò Rẹ̀

Ṣàfikún TTS nínú ohun-iní rẹ̀ nípa lórí àwọn àwọn àwòrán tí a fi pamọ́ fún ara wa tàbí REST API wa. Àwọn àwòrán gbogbo ní a lè lò nípa ìmọ̀ràn àti àwọn ẹ̀yàn.

Àwọn ìṣàfarawé ìṣẹ̀dà

Gbogbo àwọn módè́lì lórí TTS.ai lo àwọn ìwé-aládàrọ̀ àìfilọ́lẹ̀ tí a ṣí

Àwọn ìṣàmúlò-ètò	Àwọn Ààyè-iṣẹ́	Àwọn Àkọ́gbègbè
Kokoro	Apache 2.0	Tí a fẹ́
Piper	MIT	Àwọn ìkúndùǹ
VITS	MIT	Àwọn ìkúndùǹ
MeloTTS	MIT	Àwọn ìkúndùǹ
Chatterbox	MIT	Àwọn ìkúndùǹ
Tortoise TTS	Apache 2.0	Tí a fẹ́
StyleTTS 2	MIT	Àwọn ìkúndùǹ
OpenVoice	MIT	Àwọn ìkúndùǹ
Sesame CSM	Apache 2.0	Tí a fẹ́
Orpheus	Llama 3.2	"Built with Llama"

Try These Models Free

Àwọn Ìṣàfilọ́lẹ̀ Rẹ́

Rọ́ọ̀nù àwọn àwòrán fún ara rẹ̀ tàbí fi wà lórí àwọn ààyè-iṣẹ́

Ààyè-iṣẹ́ Fún Rẹ̀ Rẹ̀

Àwọn móòdù gbogbo nípa TTS.ai ní àwọn tí a lè lò nípa ìṣàfilọ́lẹ̀ àwọn ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀ nípa GitHub tàbí Hugging Face. Ṣàfihàn àwọn ìṣàfilọ́lẹ̀, àwọn ìpàrárà, àtí ìṣàfilọ́lẹ̀ àwọn ìṣàfilọ́lẹ̀ nípa GPUs rẹ̀. Ò ní ìdáràn nípa ìṣàfilọ́lẹ̀, ìdáràn, àtí ìṣàfilọ́lẹ̀.

Àwọn ààyè-iṣẹ́ ìdáràn - àwòrán kò fi àwọn sáà rẹ̀ kù
Kò ní àwọn owó nípa àgbékalẹ̀ ìṣàmúlò-ètò
Àwọn ìṣàmúlò-ètò ìṣàfarawé fún àwọn ààyè-iṣẹ́ rẹ̀
Ní ìṣàmúlò-ètò GPU (NVIDIA tí a fọwọ́sì)
O lè ṣakoso àwọn ìṣàmúlò-ètò, ìṣàmúlò-ètò, àti àwọn ìpéwọ̀n

Lo TTS.ai Hosted API

Gbaawọn ìwọlé nígbà kan sí gbogbo àwọn àwọn àwọn ààyè-iṣẹ́ 20+ nípa API REST kan. A tí n ṣakosó ìṣàfilọ́lẹ̀ GPU, àwọn ìṣàfilọ́lẹ̀ àwọn ààyè-iṣẹ́, àwọn ìṣàfilọ́lẹ̀ àwọn àwọn àwọn àwọn àwọn àwọn. Bọ́tìnì API kan náà ǹfí ọ̀kan lọ́wọ́ lọ́wọ́ lọ́wọ́ lọ́wọ́.

Kò ní àwọn ìrísí-lẹ́tà GPU tí a fẹ́
Gbogbo 20+ awọn awoṣe nipasẹ API kan
Àwọn ìṣàmúlò-ètò àti àwọn ìdárànṣe àwọn ìṣàmúlò-ètò
99.9% uptime pẹlu infrastructura redundant
Fi owò lọ́wọ́lọ́wọ́ fun ohun tí o lò

Use Hosted API Instead

Ìṣàfilọ́lẹ̀ Àìpẹ̀: API tàbí Òjútó Rẹ́

Ló àwọn API ààyè-iṣẹ́ wà, tàbí fi Kokoro hàn nínú àwọn àkókò

Option 1: TTS.ai Hosted API Ójútó

import requests

response = requests.post("https://api.tts.ai/v1/tts", json={
    "text": "Open source TTS with a simple API.",
    "model": "kokoro",
    "voice": "af_heart",
    "format": "wav"
}, headers={"Authorization": "Bearer YOUR_API_KEY"})

with open("output.wav", "wb") as f:
    f.write(response.content)

Àwọn Àtòjọ-ẹ̀yàn 2: Ìdáràn

# Install Kokoro locally
pip install kokoro

# Generate speech on your own GPU
import kokoro

pipeline = kokoro.KPipeline(lang_code="a")
generator = pipeline("Hello from your own server!", voice="af_heart")
for i, (gs, ps, audio) in enumerate(generator):
    kokoro.save(audio, f"output_{i}.wav")

View API Documentation

Open Source, Agbara-iṣe

API wa ti a fi pamọ́ jẹ́ ki TTS-ọ̀fẹ̀-ọ̀fẹ̀ gbàwọlé látì kò ní ìṣakoso GPUs.

Àwọn Ìjánu-ìsún

15,000 àwọn àmì-àṣírí nínú ìṣàfilọ́lẹ̀

4 open-source módèlè̀ ọ̀fẹ̀
Kò ní ìṣàmúlò-ètò fún ìlòòrò
Àwọn ìṣàmúlò-ètò tí a fọwọ́sì

Àwọn Ìṣàmúlò-ètò

500,000 àwọn àmì-ìwé/oṣu

Gbogbo àwọn módè́lì 20+ tí a ṣí
Ìṣàfarawé àwọn àmì-ìwé
Àwọn Ìṣàmúlò-ètò

Àwọn Àwọn Àwọn

$29

2,000,000 àwọn àmì-ìwé/oṣu

Àwọn ìṣàmúlò-ètò GPU ìṣàfarawé
Àwọn ìṣàmúlò-ètò ìpele-òkè
Àwọn ìṣàfilọ́lẹ̀ àwọn ìṣẹ̀dá

Wó Àwọn Ìtàn

Àwọn Àtòjọ-ẹ̀yàn

Àwọn ibeere àwọn ìṣàmúlò-ètò àwọn àkọlé àwọn ìṣàmúlò-ètò àìṣiró

Yes. Every model on TTS.ai uses a permissive open-source license — either MIT or Apache 2.0. We specifically exclude models with restrictive licenses (like Coqui's CPML or non-commercial CC-BY-NC). You can verify each model's license on its GitHub repository.

Àwọn àwọn ìṣẹ̀dá àìfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣàfilọ́lẹ̀-ìṣa

Òyà. Módélù fún gbogbó àwọn lè jẹ́ ààyè fún ara wọn. Klọ́ọ̀nì àwọn àwọn àwọn àwọn ààyè fún ara wọn láti inú GitHub, fi àwọn ìpèwọ̀n hàn, gba àwọn ìṣàmúlò-ètò àwọn módélù láti inú, àti láti rọ́ọ̀nù ìṣàmúlò-ètò. À tì fi àwọn àkọsílẹ̀ fún àwọn ìṣàmúlò-ètò fún ara wọn fún módélù fún ara wọn pamọ́, àti GPU, RAM, àti ìṣàfihàn Python.

Àwọn ìṣàmúlò-ètò yatọ láti inú àwọn ìṣàmúlò-ètò. Piper kò niló GPU (CPU nikan). Kokoro àti MeloTTS niló 1-2GB VRAM. Àwọn ìṣàmúlò-ètò ìṣàmúlò-ètò ìṣàmúlò-ètò ìṣàmúlò-ètò 4GB VRAM. Tortoise àti Sesame CSM niló 8GB. A NVIDIA RTX 3060 (12GB) lè rọ́ọ̀nù àwọn ìṣàmúlò-ètò ìṣàmúlò-ètò kíì.

Ya. Ààyè ìṣàmúlò-ètò tí a ṣí gbá ìṣàmúlò-ètò láti inú àwọn ìṣàmúlò-ètò ìṣàfarawé. Àwọn ìṣàmúlò-ètò bíi GPT-SoVITS àti Bark ń fi àwọn ìsàmúlò-ètò ìṣàmúlò-ètò ìṣàmúlò-ètò pamọ́. O lè kọ́ àwọn ìṣàmúlò-ètò ní pàtó àwọn ààtò ìsàlẹ̀-ètò rẹ̀ láti ṣẹ̀dá àwọn ìsàlẹ̀-ètò ìṣàfihàn tàbí lágbára ìṣàfihàn fún àwọn ètò kan.

Top open-source models (Kokoro, StyleTTS 2, Chatterbox) now match or exceed commercial services like ElevenLabs and Google TTS in quality benchmarks. The main advantage of commercial services is managed infrastructure and support, not audio quality.

A ti yọ wọn kuro. XTTS/XTTS-v2 (Coqui's CPML — kò ní ìṣowó), F5-TTS (CC-BY-NC — kò ní ìṣowó), àti Higgs-v2 (Boson License — tí a fi hàn) tí a ti yọ gbogbo wọn kuro. Módè́ẹ̀lì fún TTS.ai ní a tì ṣàyẹwo pé ò jẹ́ ìlò ní ìṣowó.

Yes. Most models accept community contributions via GitHub. You can submit bug reports, voice recordings for new languages, code improvements, and documentation. Check each model's GitHub repository for contribution guidelines and active issues.

Fi àwọn àwọn àwòrán paṣẹ àti paṣẹ nígbà tí a kò bá ṣè nípa ìrànwọ́ GPU. Àwọn àwọn ààtò GPU wà n ṣiṣẹ́ àwọn àwòrán 20+ nípa 4x Tesla P40 (96GB VRAM tó kùnà) nípa ìṣàfilọ́lẹ̀ dínámì. Fun àwọn àwọn ayẹyẹ-ara, GPU 24GB kan lè fi àwọn àwòrán 3-5 hàn nípa ìdára.

Àwọn móòdù fún àwọn àwòrán Docker àti àwọn fáìlì Docker. Fún ìṣàmúlò-ètò àwọn móòdù àwọn, o lè kọ́ àwọn ìṣàfihàn Docker ààyè-iṣẹ́ láti lò láti gba ìṣàfihàn GPU. Àwọn ààyè-iṣẹ́ àwọn sáà API wa lè lò láti fi àwọn ìṣàfihàn àwọn ìṣàfihàn pamọ́.

Most models require Python 3.10-3.12. Coqui TTS (VITS) specifically needs Python 3.11. We recommend Python 3.12 for most models. Check each model's requirements.txt for exact version compatibility.

Ya. Àwọn ìṣẹ̀dá MIT ati Apache 2.0 gba ìlòjónú ìṣowo. O lè kọ̀ àwọn àwọn ohun-ini SaaS, àwọn ìṣàmúlò-ètò móbílì, àwọn ere, àti àwọn iṣẹ́ láti ló àwọn módè́lì yìí láti mú àwọn àwọn owó ìṣẹ̀dá, àwọn ẹ̀yàn, àti àwọn ìṣàfilọ́lẹ̀ (nípá tí ìṣàfilọ́lẹ̀ jẹ́ ayọ̀).

5.0/5 (1)

Wá Open Source TTS ní ọjọ́ yìí

20+ awọn awoṣe orisun-iṣilọ, gbogbo wọn ni iwe-aṣẹ iṣowo. Lo API wa tabi olori-ararẹ - yiyan jẹ rẹ.

Ṣàfihàn Wó Àwọn Ìtàn

Àwọn àwòrán àkọlé ìṣàmúlò-ètò

_Yanju

O fẹ́ TTS.ai? Fì sọ̀kalẹ̀ fún àwọn ọrẹ̀ rẹ̀!

Àwọn Ìfànẹ́ Ìṣàmúlò-ètò TTS

Gbogbo àwọn ààyè-iṣẹ́ àìfilọ́lẹ̀

Apache

Àwọn Ààyè-iṣẹ́

GPU tí a fi pamọ́

Àwọn Àwọn Àwọn Àwọn Àwọn

Àwọn Ìṣàmúlò-ètò

Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn

Kokoro

Piper

VITS

Bark

Tortoise TTS

OpenVoice

Bii o ṣe le lo TTS orisun-aṣiṣii

Wá Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn Àwọn

Wá nínú Awáròyìn Rẹ̀

Àwọn Ìṣàfilọ́lẹ̀ Àwọn Àwọn Ìṣàfilọ́lẹ̀

Ṣẹ̀dá Àtòjọ-ètò Rẹ̀

Àwọn ìṣàfarawé ìṣẹ̀dà

Àwọn Ìṣàfilọ́lẹ̀ Rẹ́

Ààyè-iṣẹ́ Fún Rẹ̀ Rẹ̀

Lo TTS.ai Hosted API

Ìṣàfilọ́lẹ̀ Àìpẹ̀: API tàbí Òjútó Rẹ́

Open Source, Agbara-iṣe

Àwọn Ìjánu-ìsún

Àwọn Ìṣàmúlò-ètò

Àwọn Àwọn Àwọn

Àwọn Àtòjọ-ẹ̀yàn

Gbogbo awọn awoṣe TTS.ai ni o wa ni orisun ti o ṣii?

Kini iyatọ laarin MIT ati Apache 2.0 licenties?

Ń lè fi àwọn àwọn ìṣàmúlò-ètò yìí pamọ́ sí àwọn sáà mìíràn mìíràn?

GPU wo ni mò nilò láti wẹ́ẹ̀bù àwọn àwọn ìṣàmúlò-ètò TTS?

Ń lè fi àwọn àwọn àwòrán TTS àìfilọ́lẹ̀ pamọ́?

Bawo ni awọn awoṣe TTS orisun-ṣinṣin ṣe yatọ si awọn iṣẹ iṣowo?

Àwọn ìṣàmúlò-ètò wò nínú àwọn láìsì tí mò yẹ́ kí mò dẹ́nú?

Bawo ni mo ṣe lè rọ́ọ̀nù àwọn àwọn ìṣàmúlò-ètò àwọn ìṣàmúlò-ètò lọ́wọ́lọ́wọ́ nípa àwọn sáà GPU kan?

Àwòrán Docker fún ìṣàfihàn-ara-ẹni-ní?

Àwọn ìṣàfarawé Python wo ni mò fẹ́ fún ìṣàfihàn ara mì?

Ń lè kọ́ àwọn nǹkan tí a lò nípa àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn àwọn

Wá Open Source TTS ní ọjọ́ yìí