alphacep / alphacep/vosk-api

Installation instructions: missing steps

オープン
#2,049 コメント 2 件 リアクション 0 件 担当者 0 名 GitHub で見る
主要言語
Jupyter Notebook
スター
15.1k
フォーク
1.8k
PR マージ指標
30日以内にマージされた PR はありません

説明

I visited https://alphacephei.com/vosk/install to look for the installation instructions.
I want to take the easiest approach:
```cmd
pip3 install vosk
```
That worked, great! So, I don't need all the "compile yourself" stuff.

Now, let's transcribe my meeting:
```cmd
vosk-transcriber -i 260708_0212.MP3 -o Meeting.txt
```
That doesn't work, and IMHO, of course not. Because I haven't downloaded a model and I haven't specified the language to use.

The installation instructions simply skip the step of downloading a model from https://alphacephei.com/vosk/models
No worries, I downloaded a German model. That's a ZIP file. I can unzip it. But where do I need to put those files?
Let me try to help myself:
```cmd
vosk-transcriber –-help
[...]
usage: vosk-transcriber.exe [-h] [--model MODEL]
vosk-transcriber.exe: error: unrecognized arguments: –-help
```
Hmm, that's not overly helpful. What does "MODEL" stand for? Is it a path to a file, a folder, a name, a server URL?
Also, the `-i` and `-o` command line arguments are not listed. Why would the example `vosk-transcriber -i test.mp4 -o test.txt` ever work?

Honestly, it's quite hard for me to get this up and running, even for me who studied computer science. I couldn't complete this in 15 minutes and I think that is hindering the success of Vosk.

My next try:
```cmd
vosk-transcriber -i 260708_0212.MP3 -o Meeting.txt --model E:\Downloads\vosk-model-de-tuda-0.6-900k
[...]
LOG (VoskAPI:ReadDataFiles():model.cc:317) Loading CARPA model from E:\Downloads\vosk-model-de-tuda-0.6-900k/rescore/G.carpa
ERROR (VoskAPI:ReadInternal():const-arpa-lm.cc:610) ConstArpaLm section reading failed.
[...]
Exception: Failed to create a model
```

コントリビューションガイド

このリポジトリのコントリビューションガイドは索引されていません

評価

この issue はまだ評価されていません。

新しい issue をメールで受け取る

初心者向けの GitHub issue を短くまとめたダイジェスト。