Salad Transcription API: The lowest priced transcription API (from $0.10/hr)

🥗 Hey everyone! We are excited to share the beta version of Salad Transcription API with you. 1️⃣ What problem does it solve? Today's transcription APIs are massively overpriced due to custom models and expensive GPUs powering them on hyperscalers. 2️⃣ What is the Salad Transcription API? A full-suite, unified API offering everything from transcription, translation, summarization, LLM-powered insights, custom prompts and more, all for just $0.10 per hour. 3️⃣ What about the accuracy? Our benchmarks show an accuracy of 91.13%, second in the industry but at significantly less cost. 4️⃣ How is our price so low? Our API is powered by open-source Whisper large v3 running on consumer GPUs on SaladCloud, the world's largest distributed cloud. Our distributed model ensures we power the API with the lowest-priced GPUs in the market. 5️⃣ Why did we build this? Feedback from SaladCloud users showed that the market is in need for an affordable API. We were sitting on 1000s of low-cost, AI-enabled GPUs that outperformed datacenter GPUs for transcription in a benchmark. So we combined the two. Best suited for companies doing high-volume, batch transcription: - Media production companies (News agencies, podcast platforms, video hosting/production companies) - Education providers (Online platforms, universities) - Meeting/Call platforms (Customer service departments, meeting platforms,, call centers) - Speech-to-text AI tools 🙏🏻 Give the beta a try and let us know your feedback here or at cloud@salad.com.