Set the tone
Choose an emotion for a reflective passage, an excited announcement or a character's dialogue. Start with a short sample and listen to how the voice interprets it.
MiniMax Speech 2.8 HD
Give your words a voice.
Turn a script into narration worth listening to. Choose a voice, find the right emotion, and create speech in the language of your audience.
25 credits / 1,000 characters
Listen first
A quiet morning, in Russian.
Abbess · system voice · RU
Ранним утром город ещё спит. На кухне пахнет свежим кофе, а за окном медленно светлеет небо. Сегодня можно никуда не спешить: открыть любимую книгу, позвонить близкому человеку и просто побыть в этом моменте.Generated with MiniMax Speech 2.8 HD. No voice cloning needed.
One voice, different languages
Let the story travel further.
The same morning scene, read by Abbess in Russian, English, Spanish and German. Listen for yourself before choosing a voice for your next audience.
Russian
Abbess · system voice
Read the script
Ранним утром город ещё спит. На кухне пахнет свежим кофе, а за окном медленно светлеет небо. Сегодня можно никуда не спешить: открыть любимую книгу, позвонить близкому человеку и просто побыть в этом моменте.
English
Abbess · system voice
Read the script
Early in the morning, the city is still asleep. The kitchen smells of fresh coffee, and the sky outside is slowly getting brighter. Today, there is no need to rush. Open a favourite book, call someone you love, and take a moment just to be here.
Spanish
Abbess · system voice
Read the script
A primera hora de la mañana, la ciudad todavía duerme. En la cocina huele a café recién hecho y, tras la ventana, el cielo empieza a aclararse. Hoy no hace falta correr: abre tu libro favorito, llama a alguien que quieres y disfruta de este momento.
German
Abbess · system voice
Read the script
Am frühen Morgen schläft die Stadt noch. In der Küche duftet es nach frischem Kaffee, und draußen wird der Himmel langsam heller. Heute musst du dich nicht beeilen. Schlag dein Lieblingsbuch auf, ruf einen lieben Menschen an und genieße diesen Augenblick.
These are real generated samples. Pronunciation and accent can vary with the voice, language and script.
Hear the difference
The words stay. The feeling changes.
The same Russian passage, the same Abbess voice and the same speed. Only the emotion setting changes: calm, happy or sad.
The script in all three recordings
Я так долго ждала этого дня. Теперь мы снова здесь, вместе. Посмотри, как изменился наш город. Мне столько всего хочется тебе рассказать.Calm
Abbess · system voice
Happy
Abbess · system voice
Sad
Abbess · system voice
Direct the delivery
A little slower. A little more feeling.
Give the words room
Adjust speed and pitch, and use pauses to shape the rhythm. Preview names, dates and unfamiliar words before recording the whole script.
Take it into your project
Download MP3 for everyday use, or choose WAV or FLAC for your audio workflow. Sample rate, bitrate and channel settings give you control over the file.
Priced by your text
Your next script. A clear price.
Speech generation is charged by the length of your text. See the quote in the workspace before you generate. Using a system voice does not require a separate cloning step.
25credits / 1,000 characters
With welcome discount: 12.5 credits · 50% off for 14 days from your first quote
From a written thought to a finished recording.
- 01
Write your script
A video narration, a lesson, a story or a short announcement. Start with the words you want your audience to hear.
- 02
Choose the voice and delivery
Use a system voice such as Abbess, or a compatible private voice you have already cloned. Pick the language and tune the performance.
- 03
Listen, then download
Check the price before generating. Review the recording and download it for your video, podcast or presentation.
Already have a voice of your own?
Select a compatible saved MiniMax clone for your next script. To create one, upload a recording you have the right to use and confirm the required permissions. Cloning is priced separately.
Explore voice cloningBefore you press play.
Can MiniMax Speech read Russian text?
Yes. Russian is supported in Omnipix. The Russian examples on this page were generated with MiniMax Speech 2.8 HD and the Abbess system voice. Check names, stress and pronunciation in a short sample before creating a longer recording.
Are these real audio examples?
Yes. These samples were generated specifically for this page with MiniMax Speech 2.8 HD, using the Abbess system voice. The language examples use neutral delivery; the Russian emotion comparison changes only the emotion setting. The scripts are shown beside the recordings.
Do I need to clone a voice first?
No. You can start with a system voice such as Abbess. Voice cloning is a separate feature for creating a reusable private voice from a recording you are authorized to use.
Can I use one voice in different languages?
Yes. Enter your script in the target language and select the corresponding language setting. The examples on this page use Abbess in four languages. This is speech synthesis, not automatic translation: prepare the text in the language you want to hear.
Can I control emotions and pauses?
Yes. Choose an emotion, adjust speed and pitch, and add supported pause markers to the script. The result depends on the voice and text, so compare short samples before generating a full narration.
How much does speech generation cost?
The current price per 1,000 characters is shown on this page. Your generation is quoted from the text in the workspace before you start. Creating a new voice clone is a separate operation with its own price.
What will your next story sound like?
Start with a few lines. Find a voice. Make it yours.