Skip to content

Add Dramabox model#722

Open
lucasnewman wants to merge 6 commits into
Blaizzy:mainfrom
lucasnewman:add-dramabox
Open

Add Dramabox model#722
lucasnewman wants to merge 6 commits into
Blaizzy:mainfrom
lucasnewman:add-dramabox

Conversation

@lucasnewman
Copy link
Copy Markdown
Collaborator

@lucasnewman lucasnewman commented May 14, 2026

Comment thread mlx_audio/tts/generate.py

# Load reference audio for voice matching if specified
# Load reference audio for voice matching if specified. Some models own
# reference preprocessing and should receive paths unchanged.
Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is required because resampling to 48kHz (the model output rate) back to 16kHz inside the model's audio encoder results in noticeably degraded quality.

"""
if isinstance(audio_input, str):
import soundfile as sf
from mlx_audio.audio_io import read as read_audio
Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is just a drive-by fix for a soundfile import that had snuck in.

@lucasnewman lucasnewman marked this pull request as ready for review May 14, 2026 20:29
@lucasnewman lucasnewman requested a review from Blaizzy May 14, 2026 20:29
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant