Spaces:

---

openai

/

whisper

like 2.09k

Running on L40S

when i want to transcribe audio data by whisper(not files)
started from second request i get first output to
and that in each step concatenated with old data
how can i solve that problem ?