Gladia

Gladia is an AI transcription and audio intelligence provider that offers a unified API for developers. The company was founded in Paris in 2022 by CEO Jean-Louis Quéguiner and CTO Jonathan Soto. Quéguiner, formerly the Group VP in charge of AI & Quantum Computing at OVHcloud, started the company due to his frustration with existing transcription services' inability to understand his French accent and handle multilingual conversations. Soto brings over fifteen years of experience in building technical products.

The company's core product is a speech-to-text API that provides both real-time and asynchronous (pre-recorded) transcription. Its technology is engineered for high accuracy across over 100 languages and accents, low latency (under 300 milliseconds for real-time), and is designed to be free from hallucinations. A key feature is its "code-switching" capability, which allows the API to accurately transcribe conversations where speakers switch between different languages. Gladia's platform is built on proprietary models, including 'Whisper-Zero' and 'Solaria', which are optimized versions of advanced ASR (Automatic Speech Recognition) models.

Gladia's business model is B2B, serving clients in sectors like contact centers (CCaaS), sales enablement, and media by allowing them to embed transcription and audio analysis into their own products. The API is compatible with various tech stacks and protocols, including SIP, VoIP, FreeSwitch, and Asterisk. Beyond transcription, the platform offers a suite of audio intelligence add-ons such as speaker diarization (identifying who said what), summarization, sentiment analysis, word-level timestamps, and named entity recognition (NER). The company operates on a tiered pricing model that includes a free tier, a pay-as-you-go option, and custom enterprise plans.

Since its first API launch in June 2023, Gladia has grown to serve over 70,000 users and more than 600 enterprise customers, including Ausha, Circleback, Recall, and VEED.IO. The company has raised a total of $20.3 million in funding. This includes a seed round in 2023 with investors like New Wave and Sequoia Capital, and a $16 million Series A round in October 2024 led by XAnge. The company plans to use the funds to transition from a speech-to-text API to a comprehensive, end-to-end audio AI infrastructure.

Keywords: speech-to-text API, audio intelligence, AI transcription, real-time transcription, asynchronous transcription, multilingual transcription, automatic speech recognition, ASR, speaker diarization, sentiment analysis, named entity recognition, code-switching, audio analysis, voice recognition API, audio processing, natural language processing, LLM, enterprise AI, CCaaS solutions, sales enablement tools, media transcription