NNiceVois

Discord and chat audio

OGG to PTH: use chat audio to train an RVC model

Upload an OGG and start trainingDiscord clips, Telegram notes, and Vorbis or Opus exports upload directly.
Summary

Yes. An OGG uploads directly and trains an RVC .pth model. OGG usually holds Vorbis or Opus audio, which is what Discord, Telegram, and many browser recorders produce. It works - but chat codecs are tuned for low bandwidth, so the source deserves a closer listen than a studio file would.

Opus is genuinely good at speech for its size. The catch is that voice-chat applications often encode at very low bitrates, and what a call drops to stay smooth is exactly the high-frequency detail a voice model wants.

Voice training

Train an RVC voice from an audio file

Add up to 50 WAV, MP3, FLAC, M4A, or OGG files. NiceVois keeps them separate and prepares them together as one private training dataset.

.pth

This name is used for the .pth, .index, and complete ZIP package.

Standard is selected automatically.
200 epochs

Recommended: add your audio and NiceVois will suggest the right number.

Clean audio produces a better model. Use a dry voice recording with minimal music, echo, or background noise. Prepare your audio first if it needs it.

Persistent cloud storage

Your model library

Your voice models are files you own. Download them once and no platform can take them away.

Loading…
Finished model packages stay saved until you delete them.

Your .pth, .index, ZIP, and private listening preview remain available after closing the browser or restarting your computer. Full-resolution source audio and temporary training files are removed after validation.

Loading your models

Checking private cloud storage…

Listen for what chat compression removes

Before training on chat audio, play it back on headphones and listen for these:

Those artefacts are part of the training signal. The model can and will reproduce them.

Getting a usable model from chat audio anyway

If a Discord recording is what you have, it can still work. Favour longer stretches of continuous speech over many short clips, cut sections where other people talk, and avoid material recorded while the connection was poor. If the same speaker has any non-chat recording at all - a voice memo, a video soundtrack - prefer it, and consider combining sources.

What the service does with the OGG

The official RVC training workflow reads supported source audio through FFmpeg, decodes Vorbis or Opus into the representation preprocessing requires, and extracts the features used for training. NiceVois runs that workflow in a pinned cloud environment and validates the exported model before showing the downloads.

OGG versus the alternatives

QuestionOGG (Vorbis / Opus)M4A or MP3
Can this tool upload it?YesYes
CompressionLossy, often at low chat bitratesLossy, usually at higher bitrates
Typical sourceDiscord, Telegram, browser recorder, game audioPhone recorder, music library
Risk to watchVoice-chat artefacts and dropoutsRepeated re-encoding
Best choiceUse when chat audio is the only recording of the voicePrefer when a phone or file recording exists

What you receive

The finished model is available as a named .pth, a matching .index, and a ZIP containing both. The public workspace retains the completed files for the displayed seven-day window, and you can delete an inactive job from your private model library at any time.

Only have a Discord recording?

Upload the OGG directly and let the service prepare the dataset - just listen for chat artefacts first.

Train a PTH model from OGG

Technical reference: RVC Project training instructions.