Downloads · 30 days
0
reazonspeech-k2-v2
reazon-research/reazonspeech-k2-v2
reazonspeech-k2-v2 is a automatic speech recognition model from reazon-research. Use it when you need speech turned into text. The card lists the license as apache-2.0.
reazonspeech-k2-v2 is an automatic speech recognition (ASR) model trained on ReazonSpeech v2.0 corpus.
Access
Public
Updated Aug 1, 2024
Repo size
775 MB
Likes
34
Public
Repo makeup
Click a slice to open those files.
.onnx775 MB · 100%
At a glance
- Task
- Automatic Speech Recognition
- License
- apache-2.0
- Access
- Public
- Created
- Jul 28, 2024
- Updated
- Aug 1, 2024
- SHA
- 291488c8
Model card
From the Hugging Face model README
reazonspeech-k2-v2
reazonspeech-k2-v2 is an automatic speech recognition (ASR) model
trained on ReazonSpeech v2.0 corpus.
This model provides end-to-end Japanese speech recognition based on Next-gen Kaldi.
Model Architecture
-
Character-based RNN-T model. The total parameter count is 159.34M.
-
This model utilizes an enhanced Transformer architecture called Zipformer.
-
The training recipe is available on k2-fsa/icefall.
Note that this model can process Japanese audio clips up to ~30 seconds.
Usage
We recommend to use this model through our reazonspeech library.
from reazonspeech.k2.asr import load_model, transcribe, audio_from_path
audio = audio_from_path("speech.wav")
model = load_model()
ret = transcribe(model, audio)
print(ret.text)
License
- Task
- Automatic Speech Recognition
- License
- apache-2.0
- Languages
- ja
- Created
- Jul 28, 2024
- Updated
- Aug 1, 2024