Whisper returns an empty string

[maximofn]

Hello

I have installed whisper with this commands

conda create -n whisper python=3.9
conda activate whisper
conda install jupyter
conda install pytorch==1.10.1 torchvision==0.11.2 torchaudio==0.10.1 cudatoolkit=11.3 -c pytorch -c conda-forge
pip install -U openai-whisper
sudo apt update
sudo apt install ffmpeg
conda update ffmpeg
conda install chardet

I clone whisper repository with this

git clone https://github.com/openai/whisper.git
cd whisper

I test whisper with this command

whisper tests/jfk.flac --model tiny

And get this response

Detecting language using up to the first 30 seconds. Use `--language` to specify the language
Detected language: English
[00:00.000 --> 00:11.000]

It give me 5 files ls jfk.*

jfk.json  jfk.srt  jfk.tsv  jfk.txt  jfk.vtt

But it doesn't return text. If I do

cat jfk.txt

I receive nothing

With the rest of files

cat jfk.json

{"text": "", "segments": [{"id": 0, "seek": 0, "start": 0.0, "end": 11.0, "text": "", "tokens": [], "temperature": 0.0, "avg_logprob": -0.9534894824028015, "compression_ratio": 0.0, "no_speech_prob": 0.9867542386054993, "words": []}], "language": "en"}

cat jfk.srt

1
00:00:00,000 --> 00:00:11,000

cat jfk.tsv

start	end	text
0	11000

cat jfk.vtt

WEBVTT

00:00.000 --> 00:11.000

What is wrong?

Tahnks in advance

[maximofn]

I have create a new conda environment and install whisper by other way and now it works. I installed it with this commands

conda create -n whisper python=3.9
conda activate whisper
conda install jupyter
conda install pytorch torchvision torchaudio pytorch-cuda=11.6 -c pytorch -c nvidia
pip install git+https://github.com/openai/whisper.git
sudo apt update
sudo apt install ffmpeg