Gladia

Gladia is a developer-first speech-to-text API offering real-time and async transcription, diarization, translation, and summarization with transparent per-hour pricing.

Gladia (gladia.io) provides speech-to-text as an API for builders. It supports both asynchronous batch transcription and low-latency real-time streaming, along with features like speaker diarization, word-level timestamps, translation, and audio intelligence such as summarization and sentiment. Its multilingual coverage and accuracy make it a common choice for embedding transcription into meeting tools, contact centers, and media products. Pricing is transparent and usage-based, billed per hour of audio with no seat fees, and it includes a free monthly allowance plus starter credits for new accounts. Async transcription starts at a low per-hour rate, real-time is slightly higher, and volume discounts apply on scaling and enterprise tiers, including offline licensing for teams with data-residency needs. Gladia competes directly with speech-to-text APIs like AssemblyAI, Deepgram, and Speechmatics, positioning on price, speed, and European data options. It fits developers and product teams building transcription-powered features who want predictable per-hour costs and multilingual support, rather than individuals looking for a finished transcription interface.