Audio Classification
Transformers
Safetensors
unispeech-sat
emotion-recognition
speech-emotion-recognition
speech
multilingual
russian
quantized
compressed-tensors
int8
fp8
int4
Eval Results (legacy)
Instructions to use Aniemore/unispeech-sat-emotion-v1-crosslingual with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Aniemore/unispeech-sat-emotion-v1-crosslingual with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("audio-classification", model="Aniemore/unispeech-sat-emotion-v1-crosslingual")# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForAudioClassification processor = AutoProcessor.from_pretrained("Aniemore/unispeech-sat-emotion-v1-crosslingual") model = AutoModelForAudioClassification.from_pretrained("Aniemore/unispeech-sat-emotion-v1-crosslingual", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Add per-class recall and F1 across the panel; fit the banner title
Browse files
README.md
CHANGED
|
@@ -105,7 +105,8 @@ CAMEO ships no train/test partition, and the usual way to make one — a ran
|
|
| 105 |
|
| 106 |
</details>
|
| 107 |
|
| 108 |
-
|
|
|
|
| 109 |
|
| 110 |
<img src="assets/classes.svg" alt="Per-class recall and F1" width="100%">
|
| 111 |
|
|
@@ -113,6 +114,8 @@ Each card leads with the macro-F1 for that set; the rows are the detail behind i
|
|
| 113 |
|
| 114 |
Each set keeps its own class list: the spontaneous corpus has four classes and the other two have seven, and there is no correspondence between `positive` and any single one of `happiness`/`enthusiasm` to line them up with.
|
| 115 |
|
|
|
|
|
|
|
| 116 |
### Per-class recall on spontaneous speech
|
| 117 |
|
| 118 |
| class | this model | previous release |
|
|
|
|
| 105 |
|
| 106 |
</details>
|
| 107 |
|
| 108 |
+
<details>
|
| 109 |
+
<summary><b>Per class, across the panel</b> — recall and F1 for every class on every test set</summary>
|
| 110 |
|
| 111 |
<img src="assets/classes.svg" alt="Per-class recall and F1" width="100%">
|
| 112 |
|
|
|
|
| 114 |
|
| 115 |
Each set keeps its own class list: the spontaneous corpus has four classes and the other two have seven, and there is no correspondence between `positive` and any single one of `happiness`/`enthusiasm` to line them up with.
|
| 116 |
|
| 117 |
+
</details>
|
| 118 |
+
|
| 119 |
### Per-class recall on spontaneous speech
|
| 120 |
|
| 121 |
| class | this model | previous release |
|