Speech processing
Article
Speech processing is the analysis, transformation, synthesis, coding, and recognition of spoken language using signal-processing and computational methods. It treats speech as both an acoustic waveform and a carrier of linguistic information.
Typical tasks include speech recognition, speaker identification, speech synthesis, enhancement, compression, translation, diarisation, and the detection of features such as pitch, formants, phonemes, and timing.
Systems may use digital filters, spectral analysis, statistical models, machine learning, and neural networks. Performance is affected by accents, speaking style, background noise, reverberation, microphone quality, language, and the amount of training data.
Speech processing is used in telephones, voice assistants, transcription, accessibility tools, hearing devices, security, broadcasting, language learning, customer service, and human–computer interaction.
Source details and credits
- Source / publisher: Wikipedia
- URL type: WWW
- Credits: Wikipedia
- URL: https://en.wikipedia.org/wiki/Speech_processing