Paraformer (FunASR)
Non-autoregressive ASR model by Alibaba achieving fast parallel transcription.
About
Paraformer by Alibaba DAMO Academy, part of the FunASR toolkit, is a non-autoregressive speech recognition model that decodes a whole utterance in parallel rather than token by token, giving fast inference at competitive accuracy. FunASR adds VAD segmentation, punctuation, speaker diarization, and timestamps through a single API across many languages. Released under the MIT license.
Reviews (0)
Leave a Review
No reviews yet. Be the first to review!
Details
- Price
- Free
- Platform
- Local/Desktop
- Difficulty
- Intermediate (3/5)
- License
- MIT
- Minimum VRAM
- 4 GB
- Added
- Apr 3, 2026
Related Tools
Multilingual ASR model by NVIDIA supporting 4 languages with translation.
CLI tool that transcribes audio 10x faster using pipeline optimizations.
Self-supervised speech representation model by Meta for ASR.
High-performance C/C++ port of Whisper for CPU-based speech recognition.
General-purpose speech recognition model by OpenAI trained on 680K hours of multilingual audio.
JAX-based Whisper implementation optimized for TPU/GPU with 70x+ speedup.