Fully managed and continuously trained automatic speech recognition service

US10490183B2 · kind B2 · utility

Assignee

• AMAZON TECHNOLOGIES, INC. · US

Inventors

• Ashish Singh · Seattle, US
• Deepikaa Suresh · Bengaluru, IN
• Vasanth Philomin · Stolberg, DE
• Rajkumar Gulabani · Sammamish, US
• Vladimir Zhukov · Moscow, RU
• Swaminathan Sivasubramanian · Seattle, US
• Vikram Sathyanarayana Anbazhagan · Seattle, US
• Praveen Kumar Akarapu · Redmond, US
• Stefano Stefani · Issaquah, US

Key dates

• Filing date: Mar 15, 2018
• Grant date: Nov 26, 2019
• Expiry date: Mar 15, 2038

Classification

• Technology area (CPC G): Physics
• CPC primary: G10L2015/088
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

Techniques for automated speech recognition (ASR) are described. A user can upload an audio file to a storage location. The user then provides the ASR service with a reference to the audio file. An ASR engine analyzes the audio file, using an acoustic model to divide the audio data into words, and a language model to identify the words spoken in the audio file. The acoustic model can be trained using audio sentence data, enabling the transcription service to accurately transcribe lengthy audio data. The results are punctuated and normalized, and the resulting transcript is returned to the user.