We present our experience in a work-in-progress: a free-to-use speech-enabled web-based translation/post-editing workbench. We integrated spoken post-editing commands for English and French, the result of continuous feedback received through multiple user tests which will also be continued in the future. In our current version of this platform, we have two speech recognition engines that can be used: Google Web Speech and Azure speech recognition services. Speech and Post-Editing commands for English and French are integrated. Currently we design user tests on typing and spoken post-editing modes through controlled experiments. Quantitative analysis will be performed on the translated texts, speech post-editing commands used, time taken to post-edit and number of keystrokes used when necessary. The design aims to perform qualitative analysis based on questionnaires provided with a Likert Scale approach and gather user opinions as well on how to improve the workbench behavior for better usability. We aim to use our findings to improve COPECO-SPEECH for different languages and text types, add more speech commands, and let interested users try our platform for speech-based translation and post-editing