Automatic Speech Recognition
Transformers
Safetensors
Japanese
English
qwen3_asr
asr
japanese
qwen3-asr
speech
audio
Instructions to use neosophie/Qwen3-ASR-1.7B-JA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use neosophie/Qwen3-ASR-1.7B-JA with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="neosophie/Qwen3-ASR-1.7B-JA")# Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("neosophie/Qwen3-ASR-1.7B-JA") model = AutoModelForMultimodalLM.from_pretrained("neosophie/Qwen3-ASR-1.7B-JA", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -32,11 +32,26 @@ Designed for Japanese ASR tasks requiring accurate transcription of:
|
|
| 32 |
|
| 33 |
### Example Normalizations
|
| 34 |
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
|
| 38 |
-
|
|
| 39 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 40 |
|
| 41 |
## Model Details
|
| 42 |
|
|
|
|
| 32 |
|
| 33 |
### Example Normalizations
|
| 34 |
|
| 35 |
+
#### Product / Company Names
|
| 36 |
+
|
| 37 |
+
| Spoken | Expected output |
|
| 38 |
+
| ------ | --------------- |
|
| 39 |
+
| ジェミニ | `Gemini` |
|
| 40 |
+
| アンスロピック | `Anthropic` |
|
| 41 |
+
| ノートブックLM | `NotebookLM` |
|
| 42 |
+
|
| 43 |
+
#### Japanese Numerals
|
| 44 |
+
|
| 45 |
+
| Spoken | Expected output |
|
| 46 |
+
| ------ | --------------- |
|
| 47 |
+
| 一万二千三百四十五 | `12345` |
|
| 48 |
+
| 十人 | `10人` |
|
| 49 |
+
|
| 50 |
+
#### Math / Code
|
| 51 |
+
|
| 52 |
+
| Spoken | Expected output |
|
| 53 |
+
| ------ | --------------- |
|
| 54 |
+
| AイコールAプラス1 | `A=A+1` |
|
| 55 |
|
| 56 |
## Model Details
|
| 57 |
|