daveshap/keras_asr

ASR experiment using Google's Universal Sentence Encoder

34
/ 100
Emerging

This helps researchers and developers who are exploring new ways to convert spoken language into written text, particularly in experimental setups. It takes raw audio and its corresponding human-written transcriptions to create a system that can then transform new audio recordings into encoded representations and finally into text. This is designed for those working on advanced speech-to-text methodologies.

No commits in the last 6 months.

Use this if you are a researcher or AI/ML engineer experimenting with novel audio processing and natural language encoding techniques for speech recognition.

Not ideal if you need a ready-to-use, high-performance speech-to-text system for production applications without deep customization.

speech-recognition-research natural-language-processing-experimentation audio-transcription-development machine-learning-engineering AI-model-prototyping
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 13 / 25

How are scores calculated?

Stars

9

Forks

2

Language

Jupyter Notebook

License

MIT

Last pushed

Sep 26, 2018

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/daveshap/keras_asr"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.