The latest articles on DEV Community by easwee (@easwee). https://dev.to/easwee https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F3353812%2F152cf83e-af20-4958-b8fe-99b4f01e1708.jpeg https://dev.to/easwee en easwee Tue, 15 Jul 2025 10:00:22 +0000 https://dev.to/easwee/we-built-an-open-tool-to-compare-voice-apis-in-real-time-25lg https://dev.to/easwee/we-built-an-open-tool-to-compare-voice-apis-in-real-time-25lg <p>We recently built <a href="https://soniox.com/compare" rel="noopener noreferrer">Soniox Compare</a>, a tool that lets you test real-time voice AI systems side by side.</p> <p>You can:</p> <ul> <li>Speak into your mic in any language</li> <li>Upload an audio file</li> </ul> <p>The same audio is sent to multiple providers (Soniox, Google, OpenAI, etc) and their outputs appear live, side by side.</p> <p><a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fg9lusie3jcttfph7mivg.png" class="article-body-image-wrapper"><img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Farticles%2Fg9lusie3jcttfph7mivg.png" alt=" " width="800" height="491"></a></p> <p>We built this because evaluating speech APIs is surprisingly tedious. Static benchmarks often don’t reflect real-time performance, and API docs rarely cover the messy edge cases: noisy input, overlapping speech, mid-sentence language shifts, or audio from the wild.</p> <p>We wanted a quick, transparent way to test systems using the same audio under the same conditions. And to see what actually works in practice.</p> <p>So we made something simple, open, and reproducible.</p> <p><strong>Want to go deeper?</strong><br> We also open-sourced the full benchmarking framework here: <a href="https://github.com/soniox/soniox-compare" rel="noopener noreferrer">https://github.com/soniox/soniox-compare</a></p> <p>You can:</p> <ul> <li>Run batch tests on your own data</li> <li>Add new providers</li> <li>Modify the workflow to fit your use case</li> </ul> <p><strong>Try it yourself</strong></p> <ul> <li>Live tool: <a href="https://soniox.com/compare" rel="noopener noreferrer">https://soniox.com/compare</a> </li> <li>GitHub repo: <a href="https://github.com/soniox/soniox-compare" rel="noopener noreferrer">https://github.com/soniox/soniox-compare</a> </li> </ul> <p>Would love any feedback or ideas. What kinds of audio would you throw at this?</p> ai opensource webdev speechrecognition