Veloquent

Speech-to-text exists, but the output is messy. Full of "ums," run-on sentences, and awkward phrasing. You spend as much time cleaning up the transcript as you saved by not typing.

Built a desktop app that does speech-to-text with Whisper, then runs it through an AI polish layer. You can toggle polish on or off. Add presets like "casual," "formal," or "fun" to match the tone you need. The output is clean and ready to use.

• Clean, usable text from voice input

• Works in any text field on your computer

• Presets eliminate the need for manual editing

Technology Stack