What You'll Learn
- How to start, pause, and stop a recording session
- What the live transcript looks like during recording
- How to select a microphone and capture telehealth audio
- What happens if your connection drops mid-recording
- How to view your transcript and relisten to recorded sessions
- How to use dictation mode to add speech-to-text directly in a note
Before You Begin: Microphone Permissions
The first time you record, your browser asks for permission to access your microphone. Click Allow when prompted. Without this permission, recording cannot start.
If you accidentally deny permission, go to your browser's site settings (click the lock icon in the address bar) and set Microphone to Allow, then reload the page.
Choosing Your Microphone
If you have multiple audio input devices (laptop mic, headset, USB microphone), click the dropdown chevron to the right of the Start Recording button. A menu appears with two sections:
- Capture Telehealth Audio — a toggle switch at the top. When enabled, DeepCura records both your microphone and the audio playing in your browser (e.g., a patient on a video call). This feature is available on desktop browsers that support system audio capture.
- Microphone Input — a list of your detected microphones. Click one to select it. A star icon indicates your default microphone. When you select a device, it automatically saves as your new default.
Starting Your Recording
Click the Start Recording button. A brief audio disclaimer may play (your first few sessions), followed by a confirmation beep. The button area transforms into the live recording interface.
You see three controls appear:
- Pause — temporarily halts transcription. The live badge changes from red "Live" to orange "Paused."
- Stop — ends the recording session entirely and prepares the transcript for note generation.
- Volume indicator — a visual ring that pulses with your microphone input level, confirming audio is being captured.
The Live Transcript
As you speak, a transcript card appears in real time. This card includes:
- A Live badge (red pulsing dot) confirming active recording
- A recording timer showing elapsed time in HH:MM:SS format
- A scrollable transcript area where speech appears line by line as Deepgram processes it
- A note at the top: "Auto-saves every 3 min" — your transcript is periodically saved so nothing is lost if you close the tab
- A Clinical Intelligence button — opens the AI Clinical Analytics panel for real-time decision support during the encounter
- A Copy transcript button — copies the current transcript text to your clipboard
The transcript auto-scrolls to show the latest text. If you scroll up manually to review earlier content, auto-scroll pauses until you scroll back to the bottom.
Pausing and Resuming
Click Pause when you need a moment — perhaps the patient steps out or you take a phone call. The transcript header changes to an orange "Paused" badge and the timer display dims slightly. Click Resume to continue where you left off.
Pausing does not create a new transcript. Everything remains in one continuous session.
Connection Resilience
DeepCura uses a WebSocket connection for real-time transcription. If the connection drops (common when your browser tab is in the background for a while), the system:
- Continues recording audio locally as a backup
- Attempts to reconnect automatically
- Shows a notification if live text is paused but recording continues
- Processes the full audio when you stop, so no content is lost
You may see a purple "Recording" badge instead of the red "Live" badge during backup mode. This means audio is being captured even though live text has paused.
Stopping the Recording
When the encounter is complete, click Stop. The system finalizes the transcript and prepares it for note generation. The transcript text remains visible in the note thread, and you are ready to generate your AI note — covered in the next lesson.
Telehealth Recording: Capturing Both Voices
For telehealth visits, you need to capture both your voice and the patient's audio from the video platform. Follow these steps before starting the recording.
Patient audio not showing in the transcript?
If you're recording a telehealth visit but only your voice appears in the transcript, you need to enable "Capture Telehealth Audio" before recording. Click the dropdown arrow next to the microphone button, then toggle on "Capture Telehealth Audio." This must be done before you click Start Recording. Without this, DeepCura only captures your microphone — not the patient's audio from the video platform.
Step 1: Open the Patient File
Start by creating your patient or navigating to an existing patient file.
Step 2: Enable Telehealth Audio (before recording)
Click the dropdown arrow next to the microphone button to open audio settings. Toggle on "Capture Telehealth Audio" to record audio from both your mic and the browser/video platform. This is essential for capturing the patient's voice in the transcript.
Tip: You can set this as your default by clicking the star icon next to the toggle, so it's always enabled when you start recording.
Step 3: Start Recording
Click the toggle next to Start Recording.
Step 4: Select the Correct Microphone
Using wireless headphones? Be sure to select the correct input under "Microphone Input" — choose the one labeled with your device name (e.g., "Default – AirPods").
Step 5: Share Entire Screen with Audio
When the system asks what to share, choose "Entire Screen" and check the box for "Also share system audio." This is what allows the patient's voice to be captured.
Step 6: Start the Telehealth Session
With everything configured, begin your session with your patient as usual.
Step 7: Complete the Session
Once you're finished, click "Complete Session." Then choose your note template to generate a detailed summary of the visit.
Viewing the Transcript & Relisten to Recordings
After a session is complete, you can review the full transcript with timestamps and optionally replay the audio recording.
Click "Inspect Source"
To view the full transcript with timestamps behind your note and the template used to generate the note, open the note and click Inspect Source.
What You'll Find Inside
The Inspect Source view shows:
- The note template that structured your output
- The raw transcript, including timestamps for every speaker
Relisten to Your Encounters
You can save audio recordings to review any part of your session later.
To activate audio storage:
- Go to Settings > Automation Rules
- Turn on "Store Audio Recording As Backup"
To access recordings: Find them in the same location as transcripts under Inspect Source.
Dictation Mode: Speech-to-Text Inside a Note
In addition to the encounter recording mode, DeepCura offers a built-in dictation tool that lets you convert speech to text while documenting patient notes — hands-free and fast. Dictation works directly inside any open note.
Step 1: Open the Note You Want to Dictate Into
Navigate to any patient note. If you want to start from blank, click the + sign to open a prechart.
Step 2: Hover to Reveal the Editing Menu
Hover over the note you're editing. The editing menu icon bar will appear. Click anywhere in the note area to reveal the icons.
Step 3: Click the Microphone Icon
Click the microphone icon from the editing menu to launch dictation.
Step 4: Grant Microphone Access
If prompted, allow your browser or device to access the microphone to begin recording.
Step 5: Start Speaking
The dictation window opens and starts listening automatically. Speak clearly and naturally to begin transcribing.
Step 6: Insert Transcribed Text
Once finished, stop recording and click Insert Text to add the dictated content to your note.
Next Steps
Continue to Lesson 1.3 — AI Note Generation