Jun 28
[AI for Audio] Suggestion for Enhancement
Audio Translator and Captions Generator are all missing necessary Segmentation Options when exporting to captions (SRT or VTT). The minimal required options are :
- Max number of lines per segment
- Max number of words per line (in same segment)
- Max Duration Per Segment (Seconds)
- Max Characters Per Segment
- Sentence-Aware Segmentation (If enabled, the start of a new sentence will always begin a new segment)
Free transcription tools on the net already have all this, so..... ¯\_(ツ)_/¯
Planned
From Lucho: --- 1min.AI Thanks for considering this request. Truly appreciated. Also to point out that Speaker Diarization/Identification is an excellent complement which allows many useful features, to mention some : Optional Include Speaker Names in Subtitles 00:00:01.285 --> 00:00:03.804 John: Hello, I'm the Speaker1 Set VTT Subtitle Styling (bold, italic, color tags) for Voice Difference <c.white><i>Voice Over </i></c> <b>Hi, I'm Speaker1</b> Write down participants in meeting minutes and state who is in charge of each outlined task Write down who is host and guest in a podcast Export answers by speaker to table or form Transcription stats (time per speaker, clarity of speech, language used) Hopefully Speaker Diarization can be added soon to Audio Translator and Captions Generator tools 💡 Good examples of subtitle timing and AI segmentation: 🔗 ibb.co/dJ3nfzP5 | tinyurl.com/subeasy-reflow | ibb.co/23Kwbfcx