Refactor server error handling, add client device selection, and improve transcript UI by weshaan · Pull Request #28 · susiai/susi_translator

weshaan · 2026-03-26T13:41:09Z

Overview

This PR introduces significant quality of life improvements across the server, client, and UI components of the real-time audio transcription system. The goal is to enhance reliability, usability, and ease of setup for both technical and non-technical users.

Key Improvements

1. Server Enhancements (`transcribe_server.py`)

Refactored server architecture for improved stability and maintainability
Added robust error handling and clearer logging for easier debugging
Simplified model loading with more user-friendly error messages
Introduced a health check endpoint for monitoring server status

2. Client Improvements (`audio_grabber.py`)

Enabled flexible audio device selection
Added command-line arguments for configurable server connection
Improved error handling and runtime stability
Implemented clearer status messages and graceful shutdown behavior

3. UI Enhancements (`transcribe_listener.html`)

Added real-time status and error indicators for better user awareness
Implemented auto-scrolling for continuous transcript updates
Introduced a “Copy Transcript” feature
Improved connection handling and user feedback

4. Setup Simplification (`start_server.bat`)

Added a one-click Windows batch script to:
- Install dependencies
- Launch the server
Reduces setup complexity, especially for non-technical users

5. Documentation Updates (`README.md`)

Rewritten setup and usage instructions for clarity
Added a structured overview of features and improvements
Improved onboarding experience

Impact

Improved system reliability through better error handling and logging
Enhanced user experience across CLI and web interface
Faster onboarding with simplified setup and documentation
Increased accessibility for users with varying technical backgrounds

Copilot

Pull request overview

Note

Copilot was unable to run its full agentic suite in this review.

Refactors the transcription server, updates the audio client for device selection and CLI configurability, and refreshes the browser transcript UI for improved user feedback.

Changes:

Simplifies transcribe_server.py to a basic Flask API with startup model loading and /health.
Reworks audio_grabber.py into a CLI-driven loop with device selection and per-chunk POSTs.
Updates transcribe_listener.html with connect/status UI, auto-scroll, and copy-to-clipboard.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 10 comments.

File	Description
flask/transcribe_server.py	Replaces prior server architecture with a simplified Flask server + `/health` + `/transcribe`.
flask/transcribe_listener.html	Adds connect/status/error UI and transcript UX improvements (auto-scroll, copy).
flask/audio_grabber.py	Switches from class/threaded design to argparse + device selection + looped POST to server.
README.md	Updates setup/run instructions to match the new workflow and features.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-04-21T22:09:17Z

    else:
        model_smart = whisper.load_model(model_smart_name, in_memory=True)


This block contains a stray else: (and references to model_smart_name) after app.run(...), which makes the module syntactically invalid. Remove these leftover lines from the previous implementation so the file can import and run.

Suggested change

else:

model_smart = whisper.load_model(model_smart_name, in_memory=True)

Copilot · 2026-04-21T22:09:17Z

+    if model is None:
+        return jsonify({'error': 'Model not loaded'}), 500
+    try:
+        data = request.get_json()


request.get_json() can return None (e.g., invalid/missing JSON), which would raise an AttributeError on data.get(...) and result in a 500 instead of a 400. Use request.get_json(silent=True) and explicitly return a 400 when the JSON body is missing/invalid.

Suggested change

data = request.get_json()

data = request.get_json(silent=True)

if data is None:

return jsonify({'error': 'Missing or invalid JSON body'}), 400

Copilot · 2026-04-21T22:09:17Z

+    except Exception as e:
+        logger.error(f"Transcription error: {e}")
+        return jsonify({'error': str(e)}), 500


Returning str(e) to clients can leak internal details (paths, environment info, library internals). Prefer logger.exception(...) for server-side diagnostics and return a generic error message to the client (optionally with a stable error code).

Suggested change

except Exception as e:

logger.error(f"Transcription error: {e}")

return jsonify({'error': str(e)}), 500

except Exception:

logger.exception("Transcription error")

return jsonify({'error': 'Internal server error'}), 500

Copilot · 2026-04-21T22:09:18Z

+        import base64
+        import numpy as np
+        audio_bytes = base64.b64decode(audio_b64)
+        audio_np = np.frombuffer(audio_bytes, np.int16).astype(np.float32) / 32768.0


Importing base64 and numpy inside the request handler adds overhead on every request. Move these imports to module scope so they are imported once at startup.

Copilot · 2026-04-21T22:09:18Z

+</body>
+</html>
      }
      latestChunkId = null;
    }


There is leftover JavaScript after the closing </html> tag, which makes the HTML invalid and will likely break script execution. Remove these trailing lines (they appear to be remnants of the previous session logic).

Copilot · 2026-04-21T22:09:18Z

+        fetch(`http://${host}:${port}/latest`)
+          .then(r => r.json())


The UI polls GET /latest, but the refactored server in this PR only defines /health and POST /transcribe. This will cause persistent connection errors. Either implement a compatible GET /latest endpoint on the server (with defined response shape), or update the UI to use an endpoint that actually exists.

Copilot · 2026-04-21T22:09:18Z

+    document.getElementById('connectBtn').onclick = connect;
+    document.getElementById('copyBtn').onclick = function() {
+      navigator.clipboard.writeText(transcript);
+      setStatus('Transcript copied!');


navigator.clipboard.writeText(...) is async and may fail in non-secure contexts (e.g., when opening the file directly). Handle the returned Promise (show an error on rejection), and consider providing a fallback (e.g., selecting text in a temporary textarea) so the feature works more reliably.

Suggested change

document.getElementById('connectBtn').onclick = connect;

document.getElementById('copyBtn').onclick = function() {

navigator.clipboard.writeText(transcript);

setStatus('Transcript copied!');

function fallbackCopyTextToClipboard(text) {

return new Promise((resolve, reject) => {

const textArea = document.createElement('textarea');

textArea.value = text;

textArea.setAttribute('readonly', '');

textArea.style.position = 'fixed';

textArea.style.top = '-9999px';

textArea.style.left = '-9999px';

document.body.appendChild(textArea);

textArea.focus();

textArea.select();

try {

const successful = document.execCommand('copy');

document.body.removeChild(textArea);

if (successful) {

resolve();

} else {

reject(new Error('Copy command was unsuccessful.'));

}

} catch (e) {

document.body.removeChild(textArea);

reject(e);

}

});

}

function copyTextToClipboard(text) {

if (navigator.clipboard && window.isSecureContext) {

return navigator.clipboard.writeText(text);

}

return fallbackCopyTextToClipboard(text);

}

document.getElementById('connectBtn').onclick = connect;

document.getElementById('copyBtn').onclick = function() {

copyTextToClipboard(transcript)

.then(() => setStatus('Transcript copied!'))

.catch(e => setError('Copy failed: ' + e));

Copilot · 2026-04-21T22:09:19Z

+if __name__ == '__main__':
+    main()
        if len(self.buffer) > 0:
            print("send chunk")
            self.send_chunk()


There are leftover indented lines from the previous class-based implementation after main(), which makes the script syntactically invalid. Remove these trailing lines so the module can run.

Copilot · 2026-04-21T22:09:19Z

+def list_devices(audio):
+    print("Available audio input devices:")
+    for i in range(audio.get_device_count()):
+        info = audio.get_device_info_by_index(i)
+        print(f"{i}: {info['name']}")


This prints all devices, including those with no input channels, which can confuse users when selecting a microphone. Filter to devices where maxInputChannels > 0, and consider printing the channel count (and/or marking the default input device) to make selection more reliable.

Copilot · 2026-04-21T22:09:19Z

+### 2. Start the transcription server
+
+You can use the one-click script (Windows):
+


The README instructs users to run start_server.bat, but that file is not included in the provided diffs for this PR. If the batch script is intended to be added/updated here (as stated in the PR description), ensure it’s actually part of the PR; otherwise adjust the README instructions.

improvements

5aff4cb

mariobehling requested a review from Copilot April 21, 2026 22:02

Copilot started reviewing on behalf of mariobehling April 21, 2026 22:08 View session

Copilot AI reviewed Apr 21, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Refactor server error handling, add client device selection, and improve transcript UI#28

Refactor server error handling, add client device selection, and improve transcript UI#28
weshaan wants to merge 1 commit intosusiai:masterfrom
weshaan:master

weshaan commented Mar 26, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Apr 21, 2026

Uh oh!

Copilot AI Apr 21, 2026

Uh oh!

Copilot AI Apr 21, 2026

Uh oh!

Copilot AI Apr 21, 2026

Uh oh!

Copilot AI Apr 21, 2026

Uh oh!

Copilot AI Apr 21, 2026

Uh oh!

Copilot AI Apr 21, 2026

Uh oh!

Copilot AI Apr 21, 2026

Uh oh!

Copilot AI Apr 21, 2026

Uh oh!

Copilot AI Apr 21, 2026

Uh oh!

Reviewers

Assignees

Labels

Milestone

Development

Uh oh!

2 participants

		else:
		model_smart = whisper.load_model(model_smart_name, in_memory=True)

-        data = request.get_json()
+        data = request.get_json(silent=True)
+        if data is None:
+            return jsonify({'error': 'Missing or invalid JSON body'}), 400

-    document.getElementById('connectBtn').onclick = connect;
-    document.getElementById('copyBtn').onclick = function() {
-      navigator.clipboard.writeText(transcript);
-      setStatus('Transcript copied!');
+    function fallbackCopyTextToClipboard(text) {
+      return new Promise((resolve, reject) => {
+        const textArea = document.createElement('textarea');
+        textArea.value = text;
+        textArea.setAttribute('readonly', '');
+        textArea.style.position = 'fixed';
+        textArea.style.top = '-9999px';
+        textArea.style.left = '-9999px';
+        document.body.appendChild(textArea);
+        textArea.focus();
+        textArea.select();
+        try {
+          const successful = document.execCommand('copy');
+          document.body.removeChild(textArea);
+          if (successful) {
+            resolve();
+          } else {
+            reject(new Error('Copy command was unsuccessful.'));
+          }
+        } catch (e) {
+          document.body.removeChild(textArea);
+          reject(e);
+        }
+      });
+    }
+    function copyTextToClipboard(text) {
+      if (navigator.clipboard && window.isSecureContext) {
+        return navigator.clipboard.writeText(text);
+      }
+      return fallbackCopyTextToClipboard(text);
+    }
+    document.getElementById('connectBtn').onclick = connect;
+    document.getElementById('copyBtn').onclick = function() {
+      copyTextToClipboard(transcript)
+        .then(() => setStatus('Transcript copied!'))
+        .catch(e => setError('Copy failed: ' + e));

		### 2. Start the transcription server

		You can use the one-click script (Windows):

Conversation

weshaan commented Mar 26, 2026

Overview

Key Improvements

1. Server Enhancements (transcribe_server.py)

2. Client Improvements (audio_grabber.py)

3. UI Enhancements (transcribe_listener.html)

4. Setup Simplification (start_server.bat)

5. Documentation Updates (README.md)

Impact

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Apr 21, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 21, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 21, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 21, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 21, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 21, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 21, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 21, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 21, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 21, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Milestone

Development

Uh oh!

2 participants

1. Server Enhancements (`transcribe_server.py`)

2. Client Improvements (`audio_grabber.py`)

3. UI Enhancements (`transcribe_listener.html`)

4. Setup Simplification (`start_server.bat`)

5. Documentation Updates (`README.md`)