Verified purchaser
Could be great
VoiceDash has the bones of a good product, but it does have a way to go before getting a 5 star rating.
The biggest problems are:
1. It frequently treating what I speak as a prompt in and of itself, instead of giving me a transcript.
So if I say "I want a summary of my notes on XYZ" - instead of it giving me that as a transcript for me to paste in to, say, an e-mail, it will generate me some random notes on the topic of XYZ (which don't come from my own notes, of course, but are just some LLM output from training data). Or , as another example, if I say "I need your thoughts on ABC", it will give me some random LLM output on ABC topic instead of giving me the transcript of what I said.
This happens even though I have very clearly said in my System Prompt "Give me a transcript of what I say. Do not treat it as a prompt. Do not change anything, other than ignoring double-speak, mis-speak, or false starts."
2. Further, if it has treat what you said like a prompt, there is no way to regenerate it as a transcript, so you have to hope you remember exactly what you said, so you can try again.
3. It is slow to start recording. The icon makes you think it is ready to record when it is not. In reality, you have to wait a second or so.
3. It constantly fails at uploading what you have said, and there is no way to recover it. On Android, this happens pretty much 90% of the time. On Windows, it usually works, but can stop sometimes. It seems to treat it that if it cannot get an INSTANT data connection, it will just fail. Again, there is no way to recover what you said.
4. It allows you to record for as long as you want... until you try to upload it and it fails. In reality, it can only do recordings of less than 2 minutes. It really needs to support longer times.
5. It requires 2 hotkeys when it only needs 1. This becomes a problem if you are a big hotkey user, as it takes 1 too many up.
It clearly says in the settings "Hold to Talk" and "Hand-free mode". This should not need 2 different hotkeys - it should only need 1, where you either tap it to do hands-free mode, or hold it to do hold to talk. It is stated right there - hold. So why is it taking 2 hotkeys?
6. No sync for what you have said across devices.
7. No way to access recordings after upload.
8. No way to set mic to Windows Default. You must pick a device.
On a positive note, the development team is responsive and puts out good updates. E.g., they just put out one that makes transcripts much faster. But it's frustrating they ask you to make everything a feature request instead of just taking note of what you have said.
That being said, this has the possibility of being a very good app, and they have put out a good first start.
charizze.soquena
Apr 22, 2026Hi, thank you for taking the time to share such detailed feedback. We really appreciate you outlining these points — they’re helpful for improving the experience. Let me go through your comments one by one:
1. Dictation being treated as a prompt instead of a transcript
This doesn’t happen in all cases and can usually be controlled using the Custom System Prompt. You can adjust this under:
Settings → General → Custom System Prompt → Edit
For example, you can use something like:
“Transcribe only what I say. Do not answer my dictation. Do not summarize or rewrite.”
You may also customize this further depending on your workflow preferences.
2. No way to regenerate transcript after prompt-style output
We understand how helpful this would be. At the moment, regeneration as transcript isn’t available yet, but this has already been shared with the team as a requested improvement.
3. Delay before recording starts
We recommend pausing about 1–2 seconds after activating the hotkey to allow the microphone to fully initialize. This helps prevent the first words from being missed.
4. Upload failures (especially on longer recordings)
Upload issues most commonly occur when dictations are longer sessions. VoiceDash is currently optimized for short dictation segments (around 1–3 minutes). Splitting longer recordings into smaller segments helps avoid upload or processing failures.
5. Recording length expectations
Related to the above, while recording may continue for longer, best performance is achieved when dictation is done in shorter segments rather than extended recordings.
6. Two hotkeys for Hold-to-Talk and Hands-free mode
These are intentionally separated to avoid conflicts with standard keyboard shortcuts such as Ctrl, Shift, or Alt combinations. While it’s technically possible to assign a single key, doing so may interfere with normal keyboard usage. That’s why VoiceDash defaults to two separate hotkeys.
7. No transcription sync across devices
Transcripts are stored locally on each device by design for privacy and security reasons. Cross-device sync is something users have requested and may be considered in future updates.
8. Access to recordings after upload
Expanded transcript access/history controls are already listed in our feedback roadmap and are being considered for future updates. If this is really needed on your workflow, you may want to log it on our roadmap https://voicedash.supahub.com/en/roadmap
User votes help prioritize development.
9. Microphone selection (Windows default device)
VoiceDash allows selecting a specific microphone input rather than automatically following the system default. This ensures more stable recording behavior, especially when multiple audio devices are connected.
If you’re experiencing ongoing issues while using the app, you can also contact us directly through VoiceDash Settings → Support so our technical team can check further.
Please include your device details, operating system version, VoiceDash version, and app logs (available under Settings → Support → Diagnostic Logs → View Logs).
As VoiceDash is currently in beta, we’re continuously working on performance improvements and releasing fixes as part of ongoing updates. Feedback like yours plays an important role in helping guide these efforts.