Show HN: DeepSpeech based automated transcription service
32 by braindead_in | 13 comments on Hacker News.
We have been building a DeepSpeech model with our data for the past year and we have recently hit 95% accuracy on the LibriSpeech dataset. That puts us close to the published results for DeepSpeech 2. However our dataset is conversational audio and we do much better with our own internal dataset compared to PaddlePaddle. Here's a blog post on the method we followed to build our models. https://ift.tt/2N5qCNK... We have been using this internally in our service and it saves a ton of time and effort during the typing stage. It is nowhere near to the accuracy which our transcribers can achieve, but we are getting close. We are offering automated transcripts free for a limited time. Please do try it out. https://ift.tt/2yCZLpi Thanks in advance!
Subscribe to:
Post Comments (Atom)
-
Footage posted to social media shows chaotic scenes in Senegal's capital, Dakar. from BBC News https://ift.tt/4LItBfF
-
NATO Secretary General Jens Stoltenberg speaks with Senator John McCain on Capitol Hill in 2016. NATO photo CNBC: NATO is considering na... -
Russian military claims they repelled deadly incursion by Ukrainian forces on Kursk border region Moscow has said about 300 soldiers from U...
No comments:
Post a Comment