In Review
7Under consideration
Use corrections as training data
Manual corrections would be valuable data for fine-tuning an ASR model. Ideally the existing model we be automatically optimized to the user based on any corrections they make. Failing that, they could be exported for the user to use as they wish.
SRT export support
Could auto-segment into lines based on chars per line, or intelligently using a language model. Ideally, could be manually segmented with options to add colour.
Reduce frequency of new pages
The app is very quick to turn a pause for breath into a new page, to the extent that when I said one word and paused, it seemed like the app wasn't working when in fact it had just put the words on another page
Play back recording
After making a recording, add the ability to replay (and maybe also save) what you've recorded.
Pause and resume recording
When capturing in real-time, add the ability to pause and then resume transcription.
Language support
Add the option to switch to another language model. Various small (<100MB) models are available here: https://alphacephei.com/vosk/models
Offline support
Make it into a progressive web application that works offline
In Progress
Actively being built
No In Progress issues
Completed
4Recently shipped
Switch "insert" icon
The symbol looks like a green tick (esp next to the red cross) so I initially thought the app needed me to 'tick' every correct word
Hitting X doesn't remove the word
Insert button after last word
first use feedbac
I like how clean the UI is and how quickly you can get started the app is very quick to turn a pause for breath into a new page, to the extent that when I said one word and paused, it seemed like the app wasn'tworking whenin fack it had just put the words on another page I like that you can input missing words, but the symbol for the drop down looks like a green tick (esp next to the red cross) so I initially thought the app needed me to 'tick' every correct word I really like that I can type in multiple missing words at once I can't seem to remove a word even after hitting the X Need a drop down after the last word so I can add any additional text that wasn't picked up Is there a way to play back my recording so I can check what I actually said? After I stop the recording, it looks like I cannot resume it, is there a way to do that that I'm missing?