As recorded at collection: Gladia, 10 hours of speech-to-text free every month

Gladia is a speech recognition API based in France and the European Union, using the Solaria-1 model. Every account can transcribe 10 hours of audio per month on the free Starter tier. It supports both recorded audio files and real-time transcription. All core features are available on the free tier, including automatic language detection, speaker identification, word-level timestamps, and recognition of speech that mixes more than 100 languages.

Create an account on the Gladia website to get an API key (an authentication string required to use the service) right away, without registering a credit card or speaking with a sales representative. Include the key in the request header in the format `x-gladia-key: <YOUR_API_KEY>`. See docs.gladia.io for detailed instructions.

The free allowance is measured in hours of audio, not dollar credits, and usage resets each month. For comparison, AssemblyAI offers about $50 in credits (about 238 hours of recorded audio transcription, with no real-time transcription), Speechmatics offers 8 hours per month, and Deepgram offers $200 in credits. Gladia does not specify an expiration date. You can build a real-time meeting transcription tool prototype using only the free tier.