A Case Study on DSP (Speech Processing)
- 1. A Case Study on DSP Speech Processing
- 2. What is Speech Processing? 1. Speech Coding 2. Speech Recognition 3. Speech Verification 4. Speech Enhancement5. Speech Synthesis Speech processing is the application of digital signal processing (DSP) techniques to the processing and analysis of speech signals. The application of speech processing includes:
- 3. Process of Speech Production The speech production process begins when the talker formulates a message in his/her mind to transmit to listener via speech. The next step is the conversion of the message into a language code. This corresponds to converting the message into a set of phoneme sequences corresponding to sounds that make up the words, along with prosody makers denoting duration of sounds, loudness of sounds, and pitch associated with the sounds. Figure: Shows a schematic diagram of the speech production/perception process in human beings
- 4. Information Rate of the Speech Signal First Stage The discrete symbol information rate in the raw message text is rather low (about 50 bits per second corresponding to about 8 sounds per second, where each sound is one of the about 50 distinct symbols). After the language code conversion, with the inclusion of prosody information, the information rate rises to about 200 bps. Second Stage In the next stage the representation of the information in the signal becomes continuous with an equivalent rate of about 2000 bps at the neuromuscular control level and about 30000- 50000 bps at the acoustic signal level. Third Stage The continuous information rate at the basilar membrane is in the range of 30000-500000 bps, while at the neural transduction stage is about 2000 bps. Fourth Stage The higher-level processing within the brain converts the neural signals to a discrete representation, which ultimately is decoded into a low bit rate message.
- 5. Classification of Speech Sound Type 1: VOICED speech is produced when the vocal cords play an active role in the production of sound: •50: 200 Hz for male speakers •150: 300 Hz for female speakers. •200: 400 Hz for child speakers. Example: Voiced sounds (A), (E), (I). Type 2: UNVOICED Speech is produced when vocal cords are inactive. The vocal cords are held open and air flows continuously through them. Example: Unvoiced sounds (S), (F).
- 6. Formant Frequencies Speech normally exhibits one formant frequency in every 1KHz. For VOICED speech, the magnitude of the lower formant frequencies is successively larger than the magnitude of the higher formant frequencies. For UNVOICED speech, the magnitude of the higher formant frequencies is successively larger than the magnitude of the lower formant frequencies.
- 7. Basic Assumption of Speech Processing Parameters & Speech Sound 1. Phonemes: Smallest segments of speech sounds /d/ and /b/ are distinct phonemes e.g. dark and bark. 2. It is important to realize, that phonemes are abstract linguistic units and may not be directly observed in the speech signal. 3. Different speakers producing the same string of phonemes convey the same information yet sound different as a result of differences in dialect and vocal tract length and shape. 4. There are about 40 phonemes in English. 5. We can see the table for IPA (international Phonetic Alphabet) symbol for each phoneme together with sample words in which they occur.
- 8. Model for Speech Production To develop an accurate model for how speech is produced, it is necessary to develop a digital filter-based model of human speech production mechanism. The model must contain 4 steps: Steps of Speech Production Operation of the Vocal Tract Lip/Nasal Radiation Process Both Voice and Unvoiced Speech Time Frame: 10-20ms
- 9. Overall Speech Production Model
- 10. Thank You……..