Providers

Phonexia

STT 2 models no stock voices based in Czechia API via sales data from documentation synced 2026-11-18

Phonexia is a speech technology company founded in 2006 in Brno, Czech Republic, as an offshoot of the Speech@FIT research group at Brno University of Technology; it was acquired by South Korean fund Crescendo Equity Partners in 2025. Its Speech Platform 4 is on-premises software (virtual appliance with a REST API, or Docker microservices with gRPC) offering speech-to-text (its own 6th-generation engine and an enhanced Whisper-based engine), keyword spotting, language identification, speaker identification and deepfake detection. Models and licenses are supplied by Phonexia on request; no public pricing.

Speech-to-text 2

CompareModelLangsFeaturesPriceStatus
enhanced_speech_to_text_built_on_whisperEnhanced Speech to Text Built on Whisper 64 1/11 –
speech_to_textPhonexia 6th Gen Speech to Text 24 2/11 –
Esc

↑↓ move↵ open