Naturally, we LOVE this. Feature request if I may - align multiple transcripts so we can compare quality easily.
-
-
-
Thanks, that's nice to hear
At the moment, there's not really an easy way to add multiple blocks of the same type (like ner_manual), but you could do ""text": "\n\n".join(transcripts), store the character offsets of each transcript and then resolve the span offsets later. - Još 7 drugih odgovora
Novi razgovor -
-
-
Hi Ines, is there a way to import transcriber files ? I already have a lot of transcribed data that I'd like to revise :(
-
You can pre-populate the value of the input field. So if you set "field_id": "transcript" you can stream in data like this for editable transcripts: {"audio": "http://abc.com/xyz.mp3 ", "transcript": "Text..."} Basically a mix of example 3 & 4 in the forum post I linked above.
- Još 1 odgovor
Novi razgovor -
-
-
I am going to try this right away!
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
Awesome sauce!
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.
Digital native and software developer
Working on AI, Machine Learning & NLP
Founder
Developing
Documentation & discussion:
