Audio Playground — TTS and STT redesigned
New voice picker, inline settings, sample text chips, and a built-in audio player with download controls.What's new
Both Text-to-Speech and Speech-to-Text have been rebuilt from the ground up — faster, simpler, more capable.
Text-to-Speech
- Voice strip — horizontal scrollable picker showing every available voice for the selected provider
- Selected voice bar — see which voice is active, hear a sample, switch instantly
- Sample text chips — one-click buttons to load common test phrases
- Built-in audio player — controls, scrubbing, download, format selection (MP3 / WAV / Opus)
- Inline settings — model, speed, format all in one bar instead of a hidden panel
- History panel — every generation kept for the session, replay or download anytime
- Spoken text panel — see the exact text that produced the audio output
Speech-to-Text
- New header with API reference and history side panels
- Inline settings — model and language selection right next to the file uploader
- Cleaner upload flow — drag-drop or click, with progress indicator and clear error states