ServerSideHannes/las
tf 2.0 implementation of Listen, attend and spell
This project offers a foundational building block for creating speech recognition systems. It takes raw audio represented as a mel spectrogram and a sequence of previously predicted tokens (like words or phonemes), then predicts the next token in the sequence. This is a tool for machine learning engineers and researchers working on spoken language understanding.
No commits in the last 6 months.
Use this if you are developing a custom speech-to-text system and need to integrate a 'Listen, Attend and Spell' neural network architecture.
Not ideal if you need a ready-to-use speech recognition application or a pre-trained model for transcription.
Stars
21
Forks
7
Language
Python
License
—
Category
Last pushed
Jan 19, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/ServerSideHannes/las"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Compare
Higher-rated alternatives
khanld/chunkformer
ChunkFormer: Masked Chunking Conformer For Long-Form Speech Transcription
sooftware/conformer
[Unofficial] PyTorch implementation of "Conformer: Convolution-augmented Transformer for Speech...
upskyy/Squeezeformer
PyTorch implementation of "Squeezeformer: An Efficient Transformer for Automatic Speech...
WindQAQ/listen-attend-and-spell
Tensorflow implementation of "Listen, Attend and Spell" authored by William Chan. This project...
jackaduma/LAS_Mandarin_PyTorch
Listen, attend and spell Model and a Chinese Mandarin Pretrained model (中文-普通话 ASR模型)