You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Running Buzz Captions from the Mac App store v1.0.2 (137), with downloaded Whisper.cpp - Large with CoreML, on MacBook Air with 16GB memory and Sonoma 14.5.
Unable to translate German audio (mp4 files) into English (srt): A few minutes are transcribed correctly (typically 2-4 minutes) but the remainder of the lines in the transcription are the last line (or last two lines) repeated over and over again.
SRTs produced look much like the ones in 7/1/24 problem report from WeCanSee (Problem for App Store edition #818). Problem output also looks like same as that reported by nachoruiz on 3/29 and npaine01 on 4/2 in thread opened by sovvvv on 1/23/24 (Stop transcribing in the middle of the recording #673).
All mp4 files tested end with same result (varying only in small number of lines correctly produced before repeating last line(s) translated to end of file). Get same result with Whisper.cpp Medium model.
Can you provide a fix or workaround?
The text was updated successfully, but these errors were encountered:
Running Buzz Captions from the Mac App store v1.0.2 (137), with downloaded Whisper.cpp - Large with CoreML, on MacBook Air with 16GB memory and Sonoma 14.5.
Unable to translate German audio (mp4 files) into English (srt): A few minutes are transcribed correctly (typically 2-4 minutes) but the remainder of the lines in the transcription are the last line (or last two lines) repeated over and over again.
SRTs produced look much like the ones in 7/1/24 problem report from WeCanSee (Problem for App Store edition #818). Problem output also looks like same as that reported by nachoruiz on 3/29 and npaine01 on 4/2 in thread opened by sovvvv on 1/23/24 (Stop transcribing in the middle of the recording #673).
All mp4 files tested end with same result (varying only in small number of lines correctly produced before repeating last line(s) translated to end of file). Get same result with Whisper.cpp Medium model.
Can you provide a fix or workaround?
The text was updated successfully, but these errors were encountered: