Upon the request of various Scriptix customers we decided to deploy a first version of our editor! It is now possible to create subtitles and edit the results our speech recognition models generate on our platform.
Our speech recognition models apply speaker diarization when processing showing whether a male or female is speaking. With our editor you can annotate the right speakers per subtitle segment yourself, making sure the right name is added to each speaker. The search section enables you to search for a specific speaker if needed.
When creating a subtitle, you can already set specific parameters to generate output that works best for your use case such as a maximum amount of words in a subtitle segment or a maximum duration of silence between subtitle segments. With our editor you can also adjust timing of the various subtitle segments, enabling you to work out final subtitles that fit your content perfectly.
Having large files to wade through while making corrections can be frustrating. Which is why we have added a search option to our editor that enables the user to quickly navigate large files. Users can navigate based on text (speech recognition output) and speakers.
Being a B2B company, we work a lot with resellers and with our editor we can now enable their customers to make quick adjustments to the output our models generate. Resellers can add an audio track to the subtitle editor making the correction process more user friendly. Adding an audio track – for now – is only possible for resellers since we do not store any customer content by default. Additionally, we provide our resellers the possibility of redirecting their customers to our editor and have it updated with their own settings (company name, colour settings).