Automatic Speech Recognition
Transformers
Safetensors
Hindi
English
qwen3_asr
speech
speech-to-text
hindi-asr
hindi-speech-to-text
hinglish-asr
voice-agents
streaming-asr
hindi
code-switching
hinglish
maya-research
qwen3-asr
Eval Results (legacy)
Instructions to use maya-research/maya-transcribe with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use maya-research/maya-transcribe with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="maya-research/maya-transcribe")# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("maya-research/maya-transcribe") model = AutoModelForMultimodalLM.from_pretrained("maya-research/maya-transcribe", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download processor_config.json from maya-research/maya-transcribe: direct link, hf CLI and curl.
- Browser
- Download file 487 Bytes
-
https://huggingface.co/maya-research/maya-transcribe/resolve/main/processor_config.json
- Command line
-
hf download hf://maya-research/maya-transcribe/processor_config.json
-
curl -L -o processor_config.json https://huggingface.co/maya-research/maya-transcribe/resolve/main/processor_config.json
487 Bytes
| { | |
| "feature_extractor": { | |
| "chunk_length": 30, | |
| "dither": 0.0, | |
| "feature_extractor_type": "Qwen3ASRFeatureExtractor", | |
| "feature_size": 128, | |
| "hop_length": 160, | |
| "min_length": 8000, | |
| "n_fft": 400, | |
| "n_samples": 480000, | |
| "n_window": 50, | |
| "nb_max_frames": 3000, | |
| "padding_side": "right", | |
| "padding_value": 0.0, | |
| "return_attention_mask": true, | |
| "sampling_rate": 16000 | |
| }, | |
| "processor_class": "Qwen3ASRProcessor", | |
| "timestamp_segment_time": 80 | |
| } | |