Gabriele Carioli

> I tried to encode my audio into 16kHz stereo .wav with ffmpeg, but I got only `?` everywhere and no speaker number. I didn't actually do anything. I had...

I also noticed these repetitions. I did some tests (using the ggml-medium.bin model) on some audio from an episode of an Italian TV series (about 32 minutes). There are ambient...

> `whisper.cpp` does not support `.mp3` files. The input has to be 16 kHz WAV Yes, apologies, I copied the wrong output example. This one is with a 16KHz wav...

While the output on CMD terminal is correct, if you run `chcp 65001`, the bug is present (maybe just in parsing argv?) with `--output-file`, also: ``` D:\Tools\WhisperGUI>bin\main.exe --language chinese --model...

It seems like there's an attempt to fix this repetition issue in the official OpenAI-whisper: [https://github.com/openai/whisper/pull/1052](https://github.com/openai/whisper/pull/1052)

No, no newsgroup in particular. Maybe _it.test_, which is for test and has enough posts (it's full of crap). You can use _news.neodome.net_ or _freenews.netfront.net_ as a server (readonly) or...

Yes, that's also what I see. P.S. The problem with the Javascript approach is that the "subject" property is raw, while some subjects are mime encoded. And as far as...

I don't know anything about TB API and I did a simple brute-force approach. I posted a message with a subject containint XXAAXX on it.test and setup a javascript filter...

Ok, it actually works in the simplest way: ```javascript message.mime2DecodedSubject.match(/any pattern/i); ```

Don't know it this can help. I tried to force the "Subject Regex Match" filter by manually adding a rule to the existing _msgFilterRules.dat_ file in a newsgroup account directory,...