Mastodawn

DD8SF Aug 28, 2025

WebSDRs are very nice tools in #AmateurRadio :
Especially for experiments in digital modes it is quite hard to find a partner to experiment.
A SDR on 2m just enabled me to run my very first test using 16-DAPSK modulation to transmit a 1600bit/s speech signal on less than 600Hz bandwidth.
Everything worked like simulated in #gnuoctave . It looks like a good basis for further experimentation.

Show thread

DD8SF Nov 27, 2024

Today, I took some time to investigate the cause of the clicking sounds in my #M17 transmission using EnCodec and Vocos models as the codec.

After 5 minutes, I realized that I had forgotten to update the software on the receiver side. That’s why the local echo tests worked well while the transmission did not.

Here is a sample with the correct software versions running:

@m17_project #hamradio

Voice: LibriSpeech - CC BY 4.0 - https://www.openslr.org/12/

openslr.org

Open Speech and Language Resources.

Show thread

DD8SF Nov 17, 2024

Someone asked for a longer demo video of a transmission over #M17 using EnCodec/Vocos machine learning models.
So I found a longer sample and did the same recording again.

#hamradio

Voice: LibriSpeech - CC BY 4.0 - https://www.openslr.org/12/

openslr.org

Open Speech and Language Resources.

DD8SF Nov 17, 2024

I recently got the latest version of my experiment working and decided to record a demo video.

I'm using a modified version of the MVoice #M17 client to transmit frames encoded/decoded with EnCodec [1]/Vocos [2].

I achieved a high-quality real-time transmission using older CPUs: an i5-4210U for transmission and an AMD A6-8570E for reception.

@m17_project
#hamradio

[1] EnCodec: https://github.com/facebookresearch/encodec
[2] Vocos: https://gemelo-ai.github.io/vocos/
Voice: LibriSpeech - CC BY 4.0 - https://www.openslr.org/12/

GitHub - facebookresearch/encodec: State-of-the-art deep learning based audio codec supporting both mono 24 kHz audio and stereo 48 kHz audio.

State-of-the-art deep learning based audio codec supporting both mono 24 kHz audio and stereo 48 kHz audio. - facebookresearch/encodec

GitHub