Automatic Speech Recognition
Transformers
Safetensors
English
Chinese
glmasr
text2text-generation
Eval Results
Instructions to use zai-org/GLM-ASR-Nano-2512 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use zai-org/GLM-ASR-Nano-2512 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="zai-org/GLM-ASR-Nano-2512")# Load model directly from transformers import AutoProcessor, AutoModelForSeq2SeqLM processor = AutoProcessor.from_pretrained("zai-org/GLM-ASR-Nano-2512") model = AutoModelForSeq2SeqLM.from_pretrained("zai-org/GLM-ASR-Nano-2512") - Notebooks
- Google Colab
- Kaggle
Amazing!
#11
by aoiandroid - opened
It can accurately perform ASR even in relatively minor languages such as Yoruba and Afrikaans without LID. It can transcribe in that language without the user having to select the language in advance. It can also recognize Russian, Korean, Japanese, Swahili, Bengali, Arabic, Gujarati, Basque, and Zulu. This model outperforms whisper because it covers minor languages in addition to major ones. While whisper requires you to specify the source language in advance, this model works with autodetect. What's more, it has a wide language coverage. Awesome! This is the speech recognition model I've been looking for!! Thank you!