AEmotionStudio/sam-audio-models

SAM-Audio Models (BF16 Safetensors)

Ungated mirrors of Meta's SAM-Audio model weights, converted to BF16 safetensors format and redistributed under the SAM License for easier access.

What is SAM-Audio?

SAM-Audio (Segment Anything Model for Audio) is Meta AI's foundation model for isolating any sound in audio using text, visual, or temporal prompts. It can separate specific sounds from complex audio mixtures.

• Text prompts — isolate sounds by describing them (e.g. "drums", "vocals", "piano")
• Visual prompts — point at objects in video to extract their sound
• Span prompts — specify time ranges where the target sound occurs

The -tv variants are optimized for target correctness and visual prompting.

Available Models

Model	Parameters	File Size	Original
sam-audio-large-tv-bf16.safetensors	3,715,221,638	6.92 GiB	facebook/sam-audio-large-tv
sam-audio-base-tv-bf16.safetensors	1,931,243,654	3.60 GiB	facebook/sam-audio-base-tv

Files

File	Description
sam-audio-large-tv-bf16.safetensors	Large-TV model weights (BF16)
sam-audio-base-tv-bf16.safetensors	Base-TV model weights (BF16)
config.json	Model configuration
LICENSE	SAM License (required for redistribution)

Usage

# With ComfyUI-FFMPEGA (automatic download)
# Set no_llm_mode = "audio_separate" and prompt = "vocals"

# Or standalone:
from sam_audio import SAMAudio
model = SAMAudio.from_pretrained("path/to/this/repo")

License

This model is distributed under the SAM License — see the LICENSE file. Key points:

• ✅ Commercial use permitted
• ✅ Redistribution permitted (with license included)
• ✅ Derivative works permitted
• ❌ No military/warfare, nuclear, or espionage use
• ❌ No reverse engineering

Credits

• Original model by: Meta AI (FAIR)
• Paper: SAM-Audio: Segment Anything in Audio
• Redistributed by: Æmotion Studio for use with ComfyUI-FFMPEGA