Gradium

Founded by Neil Zeghidour in 2025

Convert text to speech, transcribe audio, and clone voices via API

Pricing model

Free Tier

Rating

★ People love it

Time to value

Quick Setup (< 1 hour)

You can use Gradium to build voice-enabled applications and agents using a single API. It offers text-to-speech with natural-sounding voices, speech-to-text transcription, real-time voice-to-voice translation, and voice cloning. Developers can clone a voice from a short reference recording and use it across multiple languages including English, French, German, Spanish, and Portuguese. It also supports on-device TTS for offline use and includes tools to build voice agents with minimal setup.

What Gradium does

Convert written text into spoken audio with a natural voiceTranscribe spoken audio into written textClone a voice from a reference recordingTranslate speech from one language to another in real timeRun TTS on-device without an internet connectionIntegrate voice features into an app using a REST APIBuild and deploy a voice agent from a single promptGenerate multilingual audio from a single cloned voiceText-to-speech that produces natural, expressive voices across multiple languagesSpeech-to-text transcription with high accuracyReal-time voice-to-voice translation between languagesVoice cloning from a short reference audio sampleOn-device TTS model that works offline without internetMulti-language voice cloning across English, French, German, Spanish, and PortugueseUltra-low latency voice processing suitable for real-time applicationsSingle API access to all voice AI features