Instructions

A step-by-step guide to setting up and using EchoVCast.

1. Getting Started

1

Download & Install

Download and install EchoVCast from the website.

2

Register

On first launch, register with your license key in the desktop application.

3

Download Language Packs

Select and download the language packs for the languages you want to use. Each language pack is approximately 3 GB.

2. Setting Up Audio

Microphone

Select your mic from the device dropdown in the Mic panel. The audio level meter shows input activity.

System Audio

Enable the System Audio panel to capture audio from any application — YouTube, Twitch, Discord, etc. via WASAPI loopback. Choose between Livestream and Video modes.

Each panel can be started and stopped independently.

3. Choosing Languages

  • Select your Speech Language (the language being spoken).
  • Select your Translation Language (the language to translate into).
  • Supported: English, Japanese, and Chinese (Traditional).
  • Each panel has its own independent language settings.

4. Translation

  • Toggle the Translate switch ON in each panel to enable translation.
  • Press the Start button on each panel to begin.
  • Original text appears in the top text box, translations in the bottom.
  • Gray text shows interim (in-progress) recognition.

5. OBS Integration

Text Sources

EchoVCast writes to text files (original, translated, combined) that you can add as "Read from File" text sources in OBS.

Closed Captions

Connect to OBS via WebSocket to send CEA-608 closed captions to your stream. Configure the OBS WebSocket password in Settings.

Output directory and max lines are configurable in Settings.

6. Exporting Transcripts

  • Click the export (save) button in the toolbar.
  • Export as TXT, SRT, or VTT subtitle files.
  • Exports contain the full session log with original and translated text.

7. Settings

The Models tab lets you manage language packs and switch between GPU and CPU mode. The General tab covers OBS integration, hotkeys, profanity filter, text size, and more. The Voice tab lets you tune speech pace, STT priority, VAD sensitivity, and beam size per language and audio source. The Account tab shows your license info.

8. Tips

  • Use GPU mode for best performance (NVIDIA required).
  • For best results, set STT Priority to Accuracy and Speech Pace to Fast.
  • Close other GPU-heavy apps while using EchoVCast.
  • For system audio, use Livestream mode for ongoing streams and Video mode for shorter content.
  • Use the profanity filter to keep your stream clean.
  • Pin the window to keep it always on top while streaming.
  • Best suited for streams that are not GPU-heavy or playing graphics-intensive games, as those may affect transcription performance.
  • Use your direct microphone input rather than a post-processed or virtual audio device for the best transcription accuracy.