Voice model · Deepnia
Open TTS is Deepnia’s voice model and the default choice in its voice generator. It turns up to 5,000 characters into a voice-over you direct with 22 expression tags, a 0.70×–1.20× speed range and, if you like, your own voice made from a 10–60 second recording. It is also the only Deepnia voice model that delivers a transcript and SRT captions.
The voice generator opens with Open TTS already selected. The exact cost shows before every creation; you sign up when you generate.
Put a tag right before the words it should color: 8 emotional tones such as [excited], [whispering] or [soft], and 14 audio effects such as [laughing], [sighing] or [clear throat]. The “Add tags with AI” button places them for you, 3 per sentence at most, without changing your words.
Tick “Transcript” or “SRT captions” to download the text and an SRT file timed from the word timings, estimated when they are missing. Each caption holds 8 words or 52 characters at most, a comfortable size on vertical video.
Record or upload 10–60 seconds of clear speech, confirm you have permission to use that voice, then find it in the voice library’s “Mes voix” (My voices) tab, marked as personal.
Set the speed from 0.70× to 1.20×: a little livelier for a social ad, calmer for a tutorial or an audiobook.
Search for a voice, filter by language (French, English, Spanish) or gender, and listen before you generate. The language menu offers auto-detect plus 23 languages, from Arabic to Japanese.
In the Story Audio workflow, Open TTS reads dialogues with up to 8 library voices, for a chaptered audiobook or a scene with several characters.
The “Try Open TTS” button opens Deepnia’s voice generator with the model already selected.
In the narrator library, filter by language or gender, play the voices, or choose your personal voice under “Mes voix” (My voices).
Write up to 5,000 characters, then add tags from the “Expression” panel or with “Add tags with AI”.
Choose the language and speed, tick the transcript or SRT captions under “Text files”, check the cost shown, then click “Generate voice”.
Copy a prompt, open the tool and adapt it to your project.
[excited] Three tricks for parcels people love to open! Kraft paper, a handwritten note… [whispering] and a small surprise tucked at the bottom.
In this lesson, we will build your first monthly budget together. [emphasis] Step one: list every fixed expense, from rent to your phone plan.
Welcome back to Small Shop, Big Ideas, the show for people who run their own business. [laughing] Yes, we finally fixed the microphone. Today: turning first-time buyers into regulars.
[soft] The lighthouse keeper’s house had been empty for ten years. [sighing] And yet, every night, a light came on in the top-floor window.
Bienvenido a tu nuevo plan de entrenamiento. Elige tu objetivo y te preparamos tu primera semana de ejercicios en menos de un minuto.
Chers clients, notre magasin fermera ses portes dans quinze minutes. Merci de vous diriger vers les caisses. Nous vous souhaitons une excellente soirée.
Deepnia’s other voice and music models, to pick the right one for your project.
| Model | Inputs | Settings | Best for |
|---|---|---|---|
| Open TTSThis model | Text, Authorized personal voice | Language, Speed, Expression tags | Expressive voice-overs with captions and your own voice. |
| ElevenLabs | Text | Language, Stability, Expression tags | A voice from the ElevenLabs catalogue, with three stability levels. |
| Gemini Flash TTS | Text | Expression tags | A Google voice steered by tags, with the language detected automatically. |
| MiniMax Speech 2.6 HD | Text, Authorized personal voice | Language, Speed | A cloned voice, with adjustable speed and 23 languages + Auto. |
| Deepnia Long-form | Text, PDF, Authorized personal voice | — | A whole document, text or PDF, read into a single MP3 file. |
It is Deepnia’s voice model and the default in the voice generator. It turns up to 5,000 characters into a WAV voice-over, with expression tags, adjustable speed, personal voices and an optional transcript and SRT captions.
The language menu offers auto-detect plus 23 languages, including English, French, Spanish, Arabic and Hindi, and the voice library has French, English and Spanish narrators.
Open “Text files” and tick “SRT captions” (and “Transcript” if needed): the download buttons appear under the generated voice. The SRT follows the word timings, with 8 words or 52 characters per caption at most.
Click a tag in the “Expression” panel (8 emotional tones, 14 audio effects), such as [excited] or [laughing], or let “Add tags with AI” place them: 3 per sentence at most, without rewriting your text.
Yes. Create a personal voice: record or upload 10–60 seconds of clear speech and tick the box confirming you have permission to use that voice. It then appears under “Mes voix” (My voices). Only clone a voice you have the right to use.
Yes, from 0.70× to 1.20×, with 1.00× by default.
Up to 5,000 characters per generation, delivered as one WAV file. Split longer texts into several parts.
Yes, with the Story Audio workflow: a dialogue can bring together up to 8 library voices. In the voice generator, pick one voice per generation.
The voice generator opens with Open TTS already selected. The exact cost shows before every creation; you sign up when you generate.
