● Tensorial TTS v2

Every language.
Speaking in milliseconds.

A neural voice that speaks 43 languages with two voices, plus 29 experimental ones, and starts talking before the sentence is even finished.

~70 msto the first audio in streaming mode (warm, measured on the v2 model).
~10×faster than real time on a single CPU thread. No GPU required.
43 + 29stable and experimental languages, each with both voices.
6output formats, streamed incrementally, at 22.05 or 48 kHz.
Latency

It starts talking while it is still thinking.

Text to speech is only useful in a conversation if the wait disappears. Tensorial TTS synthesizes in short windows and streams each one the moment it is ready, so the listener hears the start of a sentence while the end is still being computed.

  • Streaming, not buffering. Audio leaves in small chunks, in all six formats, with incremental encoding.
  • Light on hardware. Two seconds of speech cost about 150 ms of CPU. It runs on commodity ARM servers.
  • Built for agents and live voice. Short utterances, the common case in conversation, are exactly where the first-chunk time matters most.
Listen

Hear every language.

The same sentence in every language, synthesized by the model. Press play, switch the voice, search by name.

English en
Hello! This is Tensorial, a voice that speaks your language.
Português (Brasil) pt-BR · Portuguese (Brazil)
Olá! Esta é a Tensorial, uma voz que fala a sua língua.
Português (Portugal) pt-PT · Portuguese (Portugal)
Olá! Esta é a Tensorial, uma voz que fala a tua língua.
Español (España) es-ES · Spanish (Spain)
¡Hola! Esta es Tensorial, una voz que habla tu idioma.
Español (México) es-MX · Spanish (Mexico)
¡Hola! Esta es Tensorial, una voz que habla tu idioma.
Français fr · French
Bonjour ! Voici Tensorial, une voix qui parle votre langue.
Deutsch de · German
Hallo! Das ist Tensorial, eine Stimme, die Ihre Sprache spricht.
Italiano it · Italian
Ciao! Questa è Tensorial, una voce che parla la tua lingua.
Nederlands nl · Dutch
Hallo! Dit is Tensorial, een stem die jouw taal spreekt.
Svenska sv · Swedish
Hej! Det här är Tensorial, en röst som talar ditt språk.
Íslenska is · Icelandic
Halló! Þetta er Tensorial, rödd sem talar þitt tungumál.
Suomi fi · Finnish
Hei! Tämä on Tensorial, ääni, joka puhuu sinun kieltäsi.
Eesti et · Estonian
Tere! See on Tensorial, hääl, mis räägib sinu keelt.
Lietuvių lt · Lithuanian
Sveiki! Tai Tensorial, balsas, kuris kalba jūsų kalba.
Polski pl · Polish
Cześć! To jest Tensorial, głos, który mówi w twoim języku.
Čeština cs · Czech
Dobrý den! Tohle je Tensorial, hlas, který mluví vaším jazykem.
Slovenčina sk · Slovak
Ahoj! Toto je Tensorial, hlas, ktorý hovorí vaším jazykom.
Slovenščina sl · Slovenian
Pozdravljeni! To je Tensorial, glas, ki govori vaš jezik.
Magyar hu · Hungarian
Helló! Ez a Tensorial, egy hang, amely az ön nyelvén beszél.
Română ro · Romanian
Bună! Aceasta este Tensorial, o voce care vorbește limba ta.
Ελληνικά el · Greek
Γεια σας! Αυτό είναι το Tensorial, μια φωνή που μιλά τη γλώσσα σας.
Русский ru · Russian
Здравствуйте! Это Тензориал, голос, который говорит на вашем языке.
Українська uk · Ukrainian
Вітаю! Це Тензоріал, голос, який говорить вашою мовою.
Türkçe tr · Turkish
Merhaba! Bu Tensorial, sizin dilinizi konuşan bir ses.
العربية (السعودية) ar-SA · Arabic (Saudi Arabia)
هلا! هذا تنسوريال، صوت يتكلم لغتك.
العربية (مصر) ar-EG · Arabic (Egypt)
أهلا! ده تنسوريال، صوت بيتكلم لغتك.
العربية (الإمارات) ar-AE · Arabic (UAE)
مرحبا! هذا تنسوريال، صوت يتحدث بلغتك.
עברית he · Hebrew
שלום! זהו טנסוריאל, קול שמדבר בשפה שלך.
فارسی fa · Persian
سلام! این تنسوریال است، صدایی که به زبان شما صحبت می‌کند.
اردو ur · Urdu
ہیلو! یہ ٹینسوریل ہے، ایک آواز جو آپ کی زبان بولتی ہے۔
हिन्दी hi · Hindi
नमस्ते! यह टेंसोरियल है, एक आवाज़ जो आपकी भाषा बोलती है।
বাংলা bn · Bengali
হ্যালো! এটি টেনসোরিয়াল, এমন একটি কণ্ঠ যা আপনার ভাষায় কথা বলে।
मराठी mr · Marathi
नमस्कार! हा टेन्सोरियल आहे, तुमच्या भाषेत बोलणारा आवाज.
தமிழ் ta · Tamil
வணக்கம்! இது டென்சோரியல், உங்கள் மொழியில் பேசும் குரல்.
తెలుగు te · Telugu
నమస్కారం! ఇది టెన్సోరియల్, మీ భాషలో మాట్లాడే స్వరం.
日本語 ja · Japanese
こんにちは!これはテンソリアル、あなたのことばをはなすこえです。
한국어 ko · Korean
안녕하세요! 저는 텐소리얼, 당신의 언어로 말하는 목소리입니다.
中文(普通话) zh · Mandarin Chinese
你好!这是腾索里尔,一个会说你的语言的声音。
粵語 yue · Cantonese
你好!呢個係騰索里爾,一把識講你嘅語言嘅聲音。
ไทย th · Thai
สวัสดี! นี่คือเทนเซอเรียล เสียงที่พูดภาษาของคุณ
Tiếng Việt vi · Vietnamese
Xin chào! Đây là Tensorial, một giọng nói nói được ngôn ngữ của bạn.
Bahasa Indonesia id · Indonesian
Halo! Ini Tensorial, suara yang berbicara dalam bahasa Anda.
Kiswahili sw · Swahili
Habari! Hii ni Tensorial, sauti inayozungumza lugha yako.
No language found.
Languages

Stable, and honestly experimental.

We separate what the model was trained on from what it only borrows, so you know what you are shipping.

43 stable languages

Trained and evaluated. From English, Portuguese, Spanish and Mandarin to Hindi, Arabic (three regional variants), Japanese, Swahili and Icelandic. Full text normalization: numbers, currency, units, dates and times.

29 experimental languages

From Catalan and Basque to Kazakh, Georgian, Māori and Quechua. Same full text frontend; only the model is experimental. Published separately in the API so you can opt in on purpose.

Two voices, everywhere

Juliana and Marco Antonio speak every language, stable or experimental. Japanese mixed with Latin-alphabet words is handled in one sentence.

API

A drop-in speech endpoint.

A plain HTTP API compatible with POST /v1/audio/speech. Point your existing client at it, choose a language and a voice, and stream the audio back.

  • Six formats · streamed incrementally
  • 22.05 kHz · or 48 kHz with bandwidth extension
  • Keys, quotas and a web console
# stream Portuguese speech to a file
curl https://<your-endpoint>/v1/audio/speech \
  -H "Authorization: Bearer $TENSORIAL_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "tensorial-vits-plbert-istft-43-3.0.0",
    "input": "Olá! Esta é a Tensorial.",
    "voice": "vox_a64467727c6f9ed4",
    "language": "pt-BR",
    "response_format": "mp3",
    "stream": true
  }' --output fala.mp3

Give your product a voice in any language.

Tell us what you are building and we will get you access.

Figures: latency measured on the v2 model, warm, on local CPU hardware; first-audio time in production adds network. Stable languages were checked with an automatic judge and two sentences per language; experimental languages were not reviewed by native speakers.