Open Source Text to Speech Models
MIT, Apache 2.0 — hapana yakazvimiririra lock-in, hapana kushandiswa kwezvirambidzo, hapana zvikanganiso zvemubhadharo wemubhadharo. Usavashandisa kuburikidza neiyo inochengetwa API, kana kuti uzvichengete iwe pachako pane yako yepamutemo infrastructure nekudzora kwese.
Tarisa ikozvino
Open Source TTS Mashandisiro
Nei open-source mamodheru ari kukosha kune ako mapurojekiti
Zvese zveOpen-Source Licensed
Yese modhi pa TTS.ai inoshandisa yakavhurika-chigadzirwa chitsva chitsva chemutauro. Hapana mabhokisi emavara, hapana kurambidzwa kwemutengesi, hapana mari yemutauro isiri kutarisirwa.
MIT / Apache 2. 0
Models ine license pasi peMIT kana Apache 2.0, iyo inobvumidza zvakanyanya open-source licenses. Usashandisa zvekutengesa, shandura, shandura — hapana zvirambidzo.
Self-Hostable
Dhawunirodha chero model uye shandisa pane yako hardware. Full control pamusoro pedata rako, latency, uye infrastructure. No cloud dependency required.
GPU inovandudzwa
Mamodeli akagadzirirwa NVIDIA GPUs neCUDA rutsigiro. Piper inoshanda chete paCPU. Mamodeli akawanda anodiwa 2-8GB VRAM kuti akwanise kuongorora zvakaomarara.
Kuchengetedzwa kwenharaunda
Active open-source masangano anochengeta uye kuvandudzwa izvi mamodheru. Kubatsira kufara — kutumira bugs, kuvandudzwa, uye matsva mazita pa GitHub.
Commercial Usage OK
All models allow commercial use under their licenses.Build zvigadzirwa, kutengesa sevhisi, uye kugadzira zvemari zvemukati pasina royalties kana kubhadhara kushandiswa.
Our Open Source Model Catalog
Yese model, yayo license, uye izvo zvainoita zvakanaka
Kokoro
Free
Lightweight 82M parameter model delivering studio-quality speech with blazing-fast inference.
Yakanaka kune: Apache 2.0 — yakanakisa mhando yemahara model, 82M params, nyore ku self-host
_Tarira Kokoro
Piper
Free
A fast, local neural text to speech system optimized for Raspberry Pi and embedded devices.
Yakanaka kune: MIT — CPU-only, yakakwana yeedge zvigadzirwa uye embedded self-hosting
_Tarira Piper
VITS
Free
Conditional variational autoencoder with adversarial learning for end-to-end text-to-speech.
Yakanaka kune: MIT — Foundational architecture yakashandiswa nevamwe vakawanda kumusoro mamodheru
_Tarira VITS
Bark
Standard
Transformer-based text-to-audio model that generates realistic speech, music, and sound effects.
Yakanaka kune: MIT — zvakasiyana-siyana audio generation kugona kupfuura standard TTS
_Tarira Bark
Tortoise TTS
Premium
Multi-voice text-to-speech focused on quality with autoregressive architecture.
Yakanaka kune: Apache 2.0 — yepamusoro mhando, zvakaoma kudzidza reference kutevedzera
_Tarira Tortoise TTS
OpenVoice
Premium
Instant voice cloning with granular control over style, emotion, and accent.
Yakanaka kune: MIT — open-source voice cloning ne granular style control
_Tarira OpenVoiceMaitiro Ekushandisa Open Source TTS
Usashandisa yedu inochengetwa API kana kuisa mamodheru iwe pachako
Kuongorora Open-Source Models
Tarisa yedu catalog ye 20 + open-source TTS mamodheru.Kazhinji, mamodheru ewebhusaiti anoratidzwa nechirongwa chekuzvishandira, chimiro, zviwanikwa, uye zvinodiwa zvekuzvishandira.
Kuedza mubrowser yako
Unogona kuedza chero model zvakananga pa TTS.ai pasina kuisa chero chinhu. GPU server yedu inodzora kuongorora kuti ugone kuongorora mhando usati wazvimisikidza iwe pachako.
Self-Host kana kushandisa yedu API
Clone model repos kubva kuGitHub uye shandisa munzvimbo yako, kana kushandisa yedu inochengetwa API yekugadzira. Self-hosting inopa kudzora kwakadzama; yedu API inopa yakachengetwa nharaunda.
Build Your Kugamuchirwa
Kubatanidza TTS muzvinhu zvako nekushandisa self-hosted mamodheru kana yedu REST API. All mamodheru anogona kushandiswa muzvitoro pasina mari yechikwereti kana royalties.
Kuenzanisa kweLicense
All models on TTS.ai use commercially-friendly open-source licenses
| Model | License | Kushandiswa kweKommercial | Kuchinja | Self-Host | Kutaura |
|---|---|---|---|---|---|
| Kokoro | Apache 2.0 | Inodiwa | |||
| Piper | MIT | Isingafanirwe | |||
| VITS | MIT | Isingafanirwe | |||
| MeloTTS | MIT | Isingafanirwe | |||
| Chatterbox | MIT | Isingafanirwe | |||
| Tortoise TTS | Apache 2.0 | Inodiwa | |||
| StyleTTS 2 | MIT | Isingafanirwe | |||
| OpenVoice | MIT | Isingafanirwe | |||
| Sesame CSM | Apache 2.0 | Inodiwa | |||
| Orpheus | Llama 3.2 | "Built with Llama" |
Self-Hosting vs Hosted API
Kushanda neModels iwe pachako kana kuti tirege isu kudzora michina
Self-Host paHardware yako
Every model on TTS.ai is available as an open source project on GitHub or Hugging Face. Download the weights, install the dependencies, and run inference on your own GPUs. You have full control over latency, privacy, and scaling.
- Full data privacy — audio kamwe haachazodzokera kune yako server
- Hapana per-request mari mushure mekutanga kuiswa
- Custom fine-tuning pane yako yega data
- Requires GPU hardware (NVIDIA inokurudzira)
- You manage updates, scaling, and dependencies
Usashandisa TTS.ai Hosted API
Kuwana nyore nyore kuwanikwa kweanopfuura 20 mamodheru kuburikidza neimwe REST API. Tinoona nezve GPU provisioning, mamodheru matsva, kurodha kwechikumbiro, uye kukwira kwechiyero.Imwe chete API key inokupa kuwanikwa kweese mamodheru — hapana chikonzero chekudzora kuiswa kwezvirongwa zvakasiyana.
- Hapana GPU hardware inodiwa
- All 20+ mamodheru kuburikidza neimwe API
- Automatic model updates uye kuvandudzwa
- 99.9% uptime ne redundant infrastructure
- Pay chete kwaunoshandisa
Kutanga nekukurumidza: API kana Self-Host
Usashandisa yedu inochengetwa API, kana kuisa Kokoro panzvimbo muminetsi
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)# 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")Open Source, Yakachipa Kutengesa
Isu tinochengeta API inoita kuti open-source TTS ive yakawanikwa pasina kudzora GPUs.
Free Tier
$0
15 mari pakutanga
- 4 open-source mamodheru akasununguka
- Hapana kumbobvira kushanyira webhusaiti
- Kushambadzira kushandiswa kwakabvumirwa
Starter
$9
500,000 characters/mwedzi
- All 20+ open-source mamodheru
- Kutaura
- API kuwanikwa
Pro
$29
2,000,000 characters/mwedzi
- GPU kuongorora kwekutanga
- All premium models
- Enterprise rutsigiro
Mibvunzo Inobvunzwa Kazhinji
Mabvunzo anowanzobvunzwa nezve open source yekushandura mashoko kuita mashoko
Chii chingatibatsira kuti tiite zvakanaka? Ruzivo rwako runogona kutibatsira kugadzirisa matambudziko.
Tarisa Open Source TTS Nhasi
20+ open-source mamodheru, ese akapihwa mvumo yekutengesa. Usashandisa yedu API kana kugara uchizvishandira - sarudzo yako.