COMPARISON OF THE MAIN TEXT‑TO‑SPEECH PROVIDERS FOR MB STUDIO

The TTS market is evolving quickly: new models, new pricing, new voice quality levels. Here is a clear and updated overview of the most commonly used providers, with real‑world pros and cons.

In MB STUDIO you can activate one, some or all of the following providers and therefore you can use them simultaneously

---

ELEVENLABS – The most famous and widely used

PRO

• Large voice catalogue, especially in English.

• Multilingual voices able to speak Italian and announce English or French titles in the same session.

• Three quality levels (Standard, Turbo, Professional).

• Free plan available, limited but useful for testing.

CON

• The most expensive provider, with mandatory monthly subscription.

• Some voices have extra costs.

• Non english voices may produce pronunciation errors.

• Professional quality generation is slower.

OPENAI TTS – Very affordable and full of potential

PRO

• Very low cost, pay‑as‑you‑go pricing.

• You can start with just €5, and the first €5 are free.

• Multilingual voices with good cross‑language fluency.

CON

• Pronunciation errors in non english languages.

• Output audio level is low (AGC recommended in MB STUDIO).

• Limited voice catalogue.

INWORLD TTS – The cheapest and excellent for English song announcements

PRO

• Very nice English voices, perfect for radio‑style announcements.

• Almost free: $10 credit offered at signup.

• Large selection of English voices.

• Very fast audio generation.

CON

• very few non english voices, quite robotic.

GEMINI TTS – High quality and improving fast

PRO

• Excellent voice quality in all languages.

• Cheaper than ElevenLabs (pricing not final).

• Very natural multilingual voices.

CON

• Still in preview mode, not always stable.

• Occasional pronunciation errors in non english languages.

• Credit card required.

• Current limit of 100 requests per day.

GOOGLE CLOUD TTS – The classic service, soon replaced by Gemini

PRO

• Very cheap when using Standard, Wavenet or Neural voices.

• Good quality for announcements.

• Extremely fast generation.

CON

• Complex activation (project, API key, services).

• Credit card required.

• Pricing not always clear, despite a small free quota.

WHICH PROVIDER SHOULD YOU CHOOSE?

English song announcements → Inworld / ElevenLabs

Non english announcements → Gemini / ElevenLabs

Lowest budget → OpenAI / Inworld

Maximum quality → Gemini / ElevenLabs

Easy activation → ElevenLabs

Complex but cheap activation → Google Cloud

PRACTICAL TIPS FOR MB STUDIO

• OpenAI → enable AGC.

• Gemini → use caching.

• ElevenLabs → avoid premium voices unless needed.

• Inworld → ideal for English.

• Google Cloud → great for fast, low‑cost announcements.