2. Home
  3. /
  4. Docs
  5. /
  6. KnowledgeFlow Docs
  7. /
  8. KnowledgeFlow Key Feature...
  9. /
  10. Voice Settings

Voice Settings

Summary

The Voice feature lets you use speech input and spoken responses in KnowledgeFlow. In the front end, users can open Settings and go to the Voice tab to configure options such as Interactive Mode, silence timeout, voice selection, speech rate, speech pitch, language, voice input, and auto-read responses. The feature can be shown or hidden depending on whether voice is enabled in the application, and the settings UI shows a disabled message when it is not available.

Use Case

Scenario:

A user wants to ask questions hands-free while working. They open the Voice settings, turn on Interactive Mode, choose a preferred voice, and adjust the speech rate and pitch so responses sound natural. They then use the microphone button to speak, let the system send the message after a pause, and hear the answer read aloud.

Benefits:

  • Faster, hands-free conversation flow.
  • User-controlled speech output and input preferences.
  • Real-time transcription with the option to edit before sending in manual mode.
  • Spoken responses can be read back automatically when enabled.

Walkthrough

 

Open Voice settings

  1. Click ⚙️ Settings in the top right of the chat interface.
  2. Select the 🎤 Voice tab.
  3. If Voice is disabled, the tab shows a friendly disabled message instead of the full settings UI. Contact support@leadingai.co.uk to discuss enabling.

Configure the feature

  • Turn on Interactive Mode if you want continuous, hands-free conversation.
  • Adjust the silence timeout to control how long the system waits before auto-sending after you stop speaking. The default is 3 seconds, and the guide describes a configurable range of 1 to 10 seconds.
  • Choose a voice from the available browser voices.
  • Set speech rate and speech pitch to change how the AI sounds
  • Select the language for recognition.
  • Toggle Enable Voice Input if you want the microphone button to appear.
  • Toggle Auto-Read Responses if you want the AI to speak answers aloud.
  • Toggle Auto-Send if you want speech to be sent automatically after you finish speaking.
  • Use the Test Voice button to hear a preview of the current settings.

Use the feature in chat

  1. Click the microphone button next to the Send button to start voice input.
  2. Speak naturally and watch the transcription appear in the input box.
  3. In Interactive Mode, the message sends automatically after a pause in speech.
  4. In Manual Voice Input mode, click the microphone again to stop, review the text, and send it yourself.
  5. If Auto-Read is enabled, the AI reads the response aloud after it replies.
  6. Use the Stop button if you want to interrupt spoken output.

Save and keep your preferences

  1. Click Save Settings after making changes.
  2. Your voice preferences are stored per user in the browser, so they can persist across sessions.
  3. Tip: If the voice sounds unnatural, try a different browser voice and adjust the speech rate slightly, as some voices sound more natural than others.

Notes

  • Voice input depends on browser support. Chrome, Edge, and Safari are the best supported browsers, while Firefox is limited to voice output only.
  • The audio is processed in the browser and only text transcriptions are saved.
  • Microphone access depends on browser permissions and HTTPS.

 

On this page