Downloads · 30 days
8
6% of all-time downloads
voiceclap-bidirlm-2.5b-lora
gijs/voiceclap-bidirlm-2.5b-lora
voiceclap-bidirlm-2.5b-lora is a feature extraction model from gijs. Use it when you need embeddings to search or compare text. It is set up for sentence-transformers. The card lists the license as apache-2.0.
A rank-16 LoRA finetune of BidirLM-Omni-2.5B-Embedding trained contrastively on the voiceclap10safe mix (got-talent, emolia, majestrino, vocal bursts, ears, expresso, vox1, vox2 — 9 datasets, ~2,909 WebDataset shards)…
All-time downloads
132
Public
Parameters
2.4B
4.9 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors4.9 GB · 100%
At a glance
- Task
- Feature Extraction
- Library
- sentence-transformers
- License
- apache-2.0
- Model type
- bidirlm_omni
- Access
- Public
- Created
- Jun 10, 2026
- Updated
- Jun 10, 2026
- SHA
- 27a25911
Base models
Model card
From the Hugging Face model README
voiceclap-bidirlm-2.5b-lora
A rank-16 LoRA finetune of BidirLM-Omni-2.5B-Embedding trained
contrastively on the voiceclap_10_safe mix (got-talent, emolia,
majestrino, vocal bursts, ears, expresso, vox1, vox2 — 9 datasets, ~2,909
WebDataset shards) for voice-emotion audio↔text retrieval.
This is the best single-model on emolia global-threshold balanced
accuracy (0.6765) from the Track I LoRA sweep using the BidirLM family
(distinct from the LCO-Omni line). Adds ensemble diversity for blends
against the LCO-3B / LCO-7B LoRA variants. The LoRA delta was re-merged
into the base safetensors via the salvage_lora_snapshot.py tool to
work around a save-path bug.
Architecture
Dual-tower-style: audio + text are processed by the BidirLM-Omni transformer with mean-pooled outputs from the inner HF model.
| Base model | BidirLM/BidirLM-Omni-2.5B-Embedding |
| Embedding dim | 2,048 (L2-normalized) |
| Audio input | 16 kHz mono FLAC, max 15s at train (20s eval) |
| Total parameters | ~2.5 B |
| Loss | symmetric InfoNCE on (audio, text) batches with gather-with-grad |
Training recipe
| Split | voiceclap_10_safe.txt (~2,909 shards · ~14 M unique samples) |
| Samples seen | 76,000 × 6 epochs = ~456k (≈ 3% of one full pass) |
| LoRA | r = 16, α = 32, dropout = 0.05, target = all-linear |
| lr / wd | 1e-4 / 0.01 (cosine, warmup = 200 steps) |
| Batch | 4 × accum 8 × 4 GH200 GPUs = effective 128 |
| Precision | bf16 |
| Best epoch | 1 (selected on emolia bal_acc) |
Evaluation
Reported numbers are for epoch 1 on the two voice-emotion benchmarks the project is built around.
emolia-bench (7,984 audio · 40-emotion binary present/absent queries)
| Metric | Value |
|---|---|
| Balanced accuracy (optimal global threshold) | 0.6765 |
| Balanced accuracy (per-emotion threshold) | ~0.69 |
| Spearman ρ | ~0.19 |
emonet-voice (12,600 voice clips · 40 emotions)
| Metric | Value |
|---|---|
| top-1 accuracy | ~0.148 |
| Spearman ρ | ~0.38 |
Use case
BidirLM-Omni is a different model class than the Qwen-based LCO line, so this model provides ensemble diversity: pairing it with the LCO LoRA snapshots improves every metric over a same-family blend. The E9 ensemble that reaches emolia per-emo 0.7157 includes this checkpoint (via the underlying lowLR variants).
Quick start
import torch
from sentence_transformers import SentenceTransformer
model = SentenceTransformer(
"gijs/voiceclap-bidirlm-2.5b-lora",
trust_remote_code=True,
model_kwargs={"torch_dtype": torch.bfloat16},
)
audio_emb = model.encode("voice_clip.flac")
text_emb = model.encode("A person speaking with sadness in their voice")
score = (audio_emb @ text_emb.T).item()
How it was built
Same recipe as
gijs/voiceclap-lco-7b-lora;
only the base model differs. Salvage was applied via
salvage_lora_snapshot.py (manual ΔW = (α/r) · B @ A merge).
Caveats
- The LoRA was trained on a tiny fraction of the corpus (~3% of one full pass).
- Per-emotion-threshold balanced accuracy includes mild eval-set leakage from threshold tuning. Use optimal-global-threshold numbers for production claims.
License
Apache-2.0 (inherits from the base model).
- Task
- Feature Extraction
- Library
- sentence-transformers
- Type
- bidirlm_omni
- License
- apache-2.0
- Languages
- en, de, fr, es, zh
- Created
- Jun 10, 2026
- Updated
- Jun 10, 2026