Live Transcription

As you record, Voicord streams your audio to our speech recognition engine and displays text in real time. You will see two types of text:

Streaming text (lighter color) — Words currently being processed. These may change slightly as the AI refines its understanding of the sentence.

Finalized text — Completed sentences that have been confirmed by the AI. These are permanent and won't change.

Voicord automatically identifies different speakers during recording and labels them as Speaker 1, Speaker 2, and so on. A speaker header appears in the transcript each time the speaker changes.

When you generate an AI summary after recording, the AI may infer real speaker names from the conversation context (e.g., if someone says "Thanks, Sarah" — Speaker 2 may be labeled as Sarah).

The transcript automatically scrolls to follow the latest text as it appears. If you scroll up to review earlier content, auto-scroll pauses so you can read without interruption.

Auto-scroll resumes automatically when you scroll back down to the bottom of the transcript.

You can switch the transcription language at any time during an active recording. This is useful when speakers switch between languages mid-conversation.

The language change takes effect immediately — new audio is processed in the selected language while previously transcribed text remains unchanged.