Synthesizer. Team Members- Abhinav Prakash Avinash Prem Kumar Koyya Neeraj Kulkarni

Transcription

1 Synthesizer Team Members- Abhinav Prakash Avinash Prem Kumar Koyya Neeraj Kulkarni Project Mentor- Aseem Kushwah Project Done under Electronics Club, IIT Kanpur as Summer Project 10. 1

2 CONTENTS Sr No Description Page no 1 Introduction 3 2 Theory 4 3 Idea of Project & Features 7 4 Hardware 8 5 Software 9 6 Results 10 7 Difficulties faced 12 8 Scope for future improvement 14 9 Acknowledgements 15 2

3 Abstract: This project aims to make a synthesizer (synth), which can produce music of different musical instruments over an octave (C4-C5). Introduction: A synthesizer (or "music" synthesizer) is an electronic musical instrument that uses one or more sound generators to create waveforms which are then processed and combined in order to generate musical sounds. Synthesizers use a number of different technologies or programmed algorithms. Among the most popular waveform synthesis techniques are subtractive synthesis, additive synthesis, wavetable synthesis, frequency modulation synthesis, phase distortion synthesis, physical modelling synthesis and sample-based synthesis. In this project, we have used a method similar to additive synthesis to produce the sound very close to that of a piano for C4-C5 octave. 3

4 THEORY- Sound from musical instruments has characteristic frequency and waveform for different notes. These characteristics are definite and can thus be treated as parameters to define or thus imitate their music. Notes of music are associated with the (definite) frequencies of music. For historical and other reasons, especially in Western music, only twelve notes of fixed frequencies are used. These fixed frequencies are mathematically related to each other, and are defined around the central note, A4 (frequency 440Hz). Note Frequency (Hz) C C # 4/D b D D # 4/E b E F F # 4/G b G G # 4/A b

5 A A # 4/B b B C An octave is the interval between one musical pitch and another with half or double its frequency. In our project we aimed at producing music only over an octave ranging from C4 (middle C) to C5. Music of the same frequency, but from different instruments can be differentiated by the waveform of the music produced. Every instrument produces different waveforms patterns, which can be analysed by the corresponding Fourier series. Fourier series decomposes any periodic function (here the waveform) into the sum of a (possibly infinite) set of simple oscillating functions, namely sines and cosines. A Chord is any set of harmonically-related notes that is heard as if sounding simultaneously (a "harmonic simultaneity"). When a mechanical musical instrument produces sound, the loudness of the sound changes over time in a way that varies from instrument to instrument. The loudness or the amplitude of the sound produced over time is characterised by the ADSR envelope 5

6 Attack time is the time taken for initial run-up of level from nil to peak. Decay time is the time taken for the subsequent run down from the attack level to the designated sustain level. Sustain level is the amplitude of the sound during the main sequence of its duration. Release time is the time taken for the sound to decay from the sustain level to zero after the key is released. About PSoC: PSoC is a Programmable System on Chip i.e. we can configure the hardware according to our requirement. Advantages over Atmega: The System Clock of PSoC 24 MHz thus enabling us to have higher frequency of the interrupts as compared to 8 MHz of Atmega. PSoC has inbuilt modules similar to those of Atmega along with additional analog blocks. e.g. - DAC, Analog Comparator. PSoC has an added advantage of having pin interrupts for all the input pins unlike Atmega. 6

7 Idea: In order to produce music: We produced analog signals (imitating the required waveform), which were finally supplied to the speakers. To obtain the analog signals (electronic): By observing the relative values of amplitude of the required waveform, we generated the corresponding digital waveform sampled at 10 khz. These digital waveforms were then fed to Digital to Analog Converter (DAC) module of PSoC to give out the required analog waveform. Features: 1. Implementing ADSR of a piano : The decay function we follow is Linear. The amplitude of the note continuously decreases and eventually fades away when the key remains pressed. The time of decay is fixed. (Approximately 3 seconds). The note that is being played stops abruptly on release of the key. 2. Chords of two notes : A chord of two notes is supported by the system which means two notes can be played simultaneously 7

8 Hardware- A. Components used- 1. PSoC (CY8C PXI) 2. PSoC Development Kit 3. Push Buttons as Piano Keys 4. Speakers 5. Power Supply B. Final Assembly: In our device we have made 13 keys which include octave (C4-C5) along with 5 sharp keys. The output from the keys is then fed to the input of PSoC. Circuit Diagram of entire instrument. 8

9 Software: The software part included producing the analog sine wave corresponding to different frequencies. The software has been developed on Cypress PSoC (Programmable System on Chip) (CY29 family).the coding has been done in C Language. The different modules used in PSoC are:- 1. Timer8 8 bit timer 2. DAC8-8 bit Digital to Analog Converter 3. Interrupts- a) Timer Interrupts. b) Pin Interrupts. The sine table of notes of different frequencies is stored in the Flash Memory of PSoC. Sine table is reference table in which values of the sine wave of corresponding frequency sampled at 10 khz are stored. The pin interrupt is used to detect which key is pressed. Then the correct sine table corresponding to the key is loaded and addition of waveforms is done if needed and finally giving the digital output. This digital output is continuously fed to DAC by using timer interrupts. DAC gives the corresponding sine wave that is fed into speakers which play the notes. 9

10 Results: The waveforms obtained were recorded on the Digital Oscilloscope. Waveform corresponding to the C4 key. Waveform corresponding to C4# key. 10

11 Waveform corresponding to D4 key. As it can be seen the waveforms obtained are fairly accurate and as expected. Waveform produced when two keys are pressed thus demonstrating the beat formation and addition of two waves 11

12 Difficulties Faced: 1. Problem with using Atmega + DAC- Initially, we tried Atmega and generated a digital signal of required sine wave out of it but eventually failed at converting the digital output into analog. The DAC0808 used failed to serve our purpose. Hence we opted to PSoC due to availability of an inbuilt DAC module. Also, the possibility of a higher clock rate in PSoC implied a better sampling rate and thus a better quality of sound. 2. Synchronising the interrupts- The interrupts check the key pressed and also write the corresponding data values to the DAC. Perhaps due to the huge code in the timer interrupts, there occurred an overflow of the interrupt stack. Hence we went for optimisation of code and made use of the available pin interrupts in PSoC (which are not available in Atmega). So the timer interrupts now have only to back up the DAC. The (pin-) interrupt is raised whenever there s a change in the state of key press (one or more). Both the interrupts were synchronised using global variables. Usage of optimal data variables like into or char instead of float etc. saved us some more time. 12

13 3. Implementation of ADSR and Chords simultaneously- Despite these improvements we couldn t implement both the features of ADSR and the chords at the same time in the given time constraint, though both were individually accomplished. The original idea of The Synthesizer was only partly accomplished. Currently it works well as a piano (whose waveforms are simple sine waves) but as for the other instruments (with further complicated waveforms), all we need is a good sampling value of the waveforms. 13

14 Scope for future improvement: Implementation of both ADSR and Chord features: Perhaps a detailed analysis of the code could help implementing both the proposed features at the same time. Addition of waveforms of other instruments: String Instruments like Guitar or Air column instrument like Clarinet, Flute etc could be added the device made thus getting close to an actual synthesizer. Better study of music and waveforms (Fourier analysis): This would help to produce music more like the actual instruments (i.e. improve the quality of sound). Inclusion of memory feature: Memory feature as in we can store whatever music we have played and play it again any time later and also play it as background music for any other tune. 14

15 Acknowledgements: We express our gratefulness to the Electronics Club and the Science and Technology Council, IIT Kanpur for giving us this opportunity to pursue the project. Our sincere thanks to the club co-ordinators- Chirag Sangani, Sumeet Kumar, Abhinav Prateek and our project mentor Aseem Kushwah for encouraging us to take up the idea and guiding us throughout the project. Special thanks to Chirag Sangani who helped us get familiar with PSoC. 15