{"id":325,"date":"2023-11-20T05:35:39","date_gmt":"2023-11-20T00:05:39","guid":{"rendered":"https:\/\/callzen.ai\/blog\/?p=325"},"modified":"2024-02-01T14:51:38","modified_gmt":"2024-02-01T09:21:38","slug":"automatic-speech-recognition-in-telephonic-speech","status":"publish","type":"post","link":"https:\/\/convozen.ai\/blog\/technical-category\/automatic-speech-recognition-in-telephonic-speech\/","title":{"rendered":"Automatic Speech Recognition in Telephonic Speech"},"content":{"rendered":"
\"\"<\/figure>\n\n\n

What is Automatic Speech Recognition (ASR)<\/strong>?<\/h2>\n\n\n\n

Telephonic conversations are one of the most effective ways of communication with customers as attention during calls is usually higher than other channels. <\/p>\n\n\n\n

Any customer-centric company tends to generate a lot of audio data, however, audio can be difficult to analyze and derive insights from. This is where Automatic Speech Recognition (ASR<\/strong>) comes in. ASR is a technology that enables computers to understand speech and convert it into textual information.<\/p>\n\n\n\n

Multilingual ASR<\/strong><\/h2>\n\n\n\n

convozen.AI serves customers situated at multiple locations across the world, and we interact in their native language. This provides another big challenge in building an understanding of different languages & accents. <\/p>\n\n\n\n

How do ASR systems work?<\/strong><\/h2>\n\n\n\n

ASR systems usually have 2 major steps <\/strong>in speech-to-text processing. <\/p>\n\n\n\n

    \n
  1. Feature extraction from audio<\/li>\n\n\n\n
  2. Mapping learned features to possible text sequences. <\/li>\n<\/ol>\n\n\n\n

    1. Feature extraction <\/strong><\/h3>\n\n\n\n

    Raw audio data consists of signals sampled at a predefined frequency. Usually, this is set at 16KHz, implying 16000 samples within 1 second. Human speech has an average of 12 characters \/ second. So compressing this information is vital to having a quality ASR system. There are several ways to extract features from raw audio.<\/p>\n\n\n\n