---
license: apache-2.0
base_model: mispeech/dasheng-base
---

# dasheng-base-env-encoder

`mispeech/dasheng-base` fine-tuned (LoRA, merged) to extract environment/background
embeddings from speech. Trained on `ChristianYang/Env-TTS-Clean` with
conversation-level AAMSoftmax. Auto-uploaded from training step **100000**.

## Usage

```python
import torch
from transformers import AutoModel, AutoFeatureExtractor
from huggingface_hub import hf_hub_download

repo = "ChristianYang/dasheng-base-env-encoder"
backbone = AutoModel.from_pretrained(repo, trust_remote_code=True).eval()
fe = AutoFeatureExtractor.from_pretrained(repo, trust_remote_code=True)
head = torch.load(hf_hub_download(repo, "head.pt"), map_location="cpu", weights_only=True)
# pipeline: 16 kHz wav -> fe -> mel -> backbone.encoder tokens [B,T,768]
# -> Conv1d x2 (head["cnn"]) -> masked attentive pooling (head["pool"]) -> 768-d embedding
```

Head weights (`head.pt`): two Conv1d(768,768,k=3) layers + attentive pooling
(Linear(768,1)). Token count = mel frames // 4.