Jérôme Louradour

RuntimeError: No SGEMM backend on CPU

2

I am testing the branch `word-level-timestamps` https://github.com/guillaumekln/faster-whisper/commit/dc780dcbe01daf9f3b082c0153c315f683ed829a I installed your fork of CTranslate2 https://github.com/guillaumekln/CTranslate2.git on branch `whisper-align`. When I try to run inference with word timestamps, I get: ``` Traceback...

Bug report - Regression of VAD quality between 3.1 and 4.0 (speech detected on perfect silence)

7

## 🐛 Bug On some audio, the quality of the VAD is reallly worse in the latest version v4.0, compared to what it was in v3.1 More precisely, v4.0 detects...

When editing a conversation, I show no way to undo the last modifications. This could be useful.

"Identification des locuteurs" or "speaker identification" refers to the process of recovering the identity of the speakers (assuming one can link the identity of a speaker to a set of...

Post-processing steps of tokenizer (string replacements) are not included in the GGUF model

2

It seems that the string replacements in the post-processing of the tokenizer are not included in the GGUF model. Hence some LLM with fancy tokenizers can have the output text...