ChordATune - A Solution for Song writing and piano music harmonization

I decided to use this blog for technical purposes (not too geeky but interesting work in the computer field) I am starting with my favorite!! one of my own creations "ChordATune"..
ChordATune is an emotion based melody/tune harmonizer focusing on piano music. One of the biggest problem in song writing or piano playing is the problem of harmonization. Most people know how to write beautiful songs o play creative tunes but they find its hard to accompany their tune with the adequate chords. This problem is addressed by ChordATune which gives you the chance to accompany your melody with chords according to your emotions, styles drum beats and tempo.

ChordATune features

• Create your own melodies using a virtual piano/ a manuscript editor.
• Generate harmony according to emotions interactively (If you don't like the harmony u can always change the emotional factor o the genre and get a complete different set of chords).
• Add Drum beats to your melody
• Arrange harmony according to the genre (user preferred style ie Disco, Rhumba, Swing, March, Rock).
• Change the tempo.
• Display guitar tabular format (guitar chords ) for a tune.
  
• Generate manuscript notation to your harmony/melody
• Save MIDI files/Manuscript files in PDF format
• Play Stop Print functionality

Check out ChordATune in action

http://www.youtube.com/watch?v=I3ZeizWDnOc

In case if you are interested in the technical background ...

ChordATune uses machine learning technology to generate the most suitable chords for a given melody... it can be called as an interactive AI system where the AI properties are created dynamically at run time according to the user input. This uses Hidden Markov Model which is a statistical mathematical model to generate the chord progression. Further, ChordATune is based of Music Technology, MIDI processing, Dynamic Programming, Multi Processing and Automatic Music Composition. Around 300 lead sheets are used to train ChordATune using heuristic and data driven approaches.

Introduction to Computer Music Harmonization

Automatic harmonization history 

Among musical compositional systems there has been a large number of researchs carried out in the field of automatic harmonization from the early 1950s onwards. When looking at automatic harmonization history the most pioneering work in automatic harmonization is that of Rothgeb who developed a SNOBOL, program to solve the harmonization problem and to identify the voice leadings notes to accompany it. Rothgeb used rule base approach that has a set of ‘if statements’ according ot the musical domain. His main aim was not focusing on automatic harmonzation but to test the computational soundness of two bass harmonization theories from the eighteenth century (Rothgeb, 1969). Afterwards Steals in 1979 proposed a system to use constraints to create passing chords as chords that could be inserted between two given chords. These passing chords must satisfy some musical constraints, such as interval relations between the roots of the first, passing, and last chords. Further, Steals used essentially a frame system, augmented with a bread-first search (Steals, 1979).

The above evidence suggests that the first works for automatic harmonization were carried by using rule base approaches by representing set of musical rules. Furthermore, this indicates that no constraint satisfaction algorithm was used and the main concern was about mastering the combinatorial explosion, by putting more knowledge in the solver.

Automatic Harmonization – AI Techniques

There have been several approaches taken for automatic harmonization in artificial neural networks (ANN). There has been a large number of research carried out in the field of music composition using neural networks available in the literature; Todd (1989, 1991) used a feed-forward ANN with feedback for melody generation, Lewis (1991), Hild et al. (1992) used a neural network to harmonize chorales that divides harmonization in to three parts 1) harmonic skeleton 2) chord skeleton and 3) ornamentation, Mozer (1991, 1994) generated melody using ANN, Stevens and Wiles (1993), Bellgard and Tsang (1994) constructed a Boltzmann machine for harmonization which generates harmonies non-deterministically. Further, Toviainen (1995) trained neural networks for jazz improvisation and later Hörnel and Degenhardt (1997) generated harmonization for four part chorals in baroque style.

The most of the automatic harmonization used neural networks are for four part choral harmonization. Automatic harmonization using neural networks, Schwanauer the developer of MUSE that harmonize chorals, claims that there are five learning techniques learning by rote, learning from failure, learning from examples, learning by analogy and learning from discovery (Schwanauer,1993).

Even though neural network approaches are widely used among musical composition systems Papadopoulos and Wiggins discusses that there are many disadvantages in neural network approach for musical compositions, in his analysis he claims that “The representation of time cannot be dealt efficiently even with ANNs that have feedback” (Papadopoulos & Wiggins, 2007). Further, he states that neural networks cannot reproduce fully trained data set. Even if they are fully trained it is not generalized (Papadopoulos & Wiggins, 2007). Toviainen (1999) claims that neural networks fail to pick up the higher-level features of music such as phrasing or tonal functions. Further, he explains that a neural network often solves toy problems, with many simplifications when compared to knowledge based approaches (Toiviainen, 1999). However, the review reveals that the neural network approaches used for musical compositions suggests that neural networks were used extensively in the past years for musical applications and they have been relatively successful especially in domains such as perception and cognition.

Automatic harmonization – Knowledge Based Systems  

The most pioneer technique used for music composition is using knowledge base concepts. Unlike neural network that learns from examples, knowledge based systems work according to the rules. There can be three types of knowledge based systems 1) rule based expert systems, Constraint logic programming and case based reasoning. There has been several approaches on automatic harmonization using rule based approach. Ebcioglu implemented a choral harmonizing system which is a rule based expert system using Backtracking Specification Language (Ebcioglu, 1988). Tsang and Aitken (1991) and Pachet and Roy (1998) used constraint logic programming (CLP) and constraint satisfaction techniques (CSP) respectively for four part choral harmonization. However, Sabater et al also developed four part automatic harmonization for vocals using rules and cased based reasoning but his approach failed due to constructing the harmonization purely on rules. Reason for his failure was because rules don’t make music but music makes rules. The advantage of using a knowledge based approach is because each newly correctly harmonized piece can be memorized and made available as a new example to harmonize other melodies which are also known as learning by experience process (Mantaras & Arcos, 2002). When analyzing the above existing work the main advantages of rule based systems are, their ability to be stand-alone programs, and their ability to explain their choice of action; furthermore, knowledge based systems can introduce explicit structures or rules. Even though knowledge based systems can be a good candidate for musical composition Papadopoulos and Wiggins claims that following disadvantages can be seen in knowledge based systems. 1) Knowledge elicitation is difficult and time consuming, 2) gaps between expert and programmer because representation is not flexible and also knowledge based systems can be too complicated with more exceptions to the rules.

Automatic harmonization – Mathematical Model

 

Mathematical model also known as Stochastic processes is a machine learning technique which is widely used in many applications and research fields such as; data mining, speech recondition, hand writing recondition and computer vision. One of the most popular mathematical models is Hidden Markov Model; it has been used successfully in genre classification, instrument identification and key estimation in the field of music systems (Levitt, 1993). There are few approaches taken for automatic harmonization using Hidden Markov Model. Allan and William designed a harmonization model for Bach chorales using Hidden Markov Models. In this model, the visible states represent melody notes and the hidden states are chords. This model has to be provided sequences of chords as input, restricting its applicability in more general settings (Allan & Williams, 2005). Hanlon and Ledlie have developed CPU Bach an automatic choral harmonization system which harmonizes melody lines for the style of composer Bach. This system breaks the harmonization process in to two parts 1) generation of a harmonic progression and 2) realization of harmonic progression into four part melody lines. Hanlon and Ledlie use HMM to model the chord progression and derive the realization of vocal lines using constraint satisfaction method. That has been successful for a large number of melodies (Hanlon & Ledlie, 2002). Papadopoulos and Wiggins claims the main draw back of this models are one must needs to find the probablities by analyzing many data therefore, lot of data is needed and also it is difficult to capture higher or more abstract levels of music (Papadopoulos & Wiggins, 2007). However, HMM has been sucessfully used in many applications as disscussed above. Furthermore, HMM is more robust and flexible compared to other models, and also its complexity is low.

Automatic harmonization – Genetic Algorithms

There have been several researches done in automatic composition using Genetic Algorithms (GA). Horner and Ayres have successfully developed a system that generates four-part harmony using genetic algorithms. However, the chords needed to be given and the system produces the four melody lines according to the chords (Horner, 1995). Phon-Amnuaisuk and Wiggins created another harmonizing system using GA in their system, soprano information is input by the users and the GA generates the other three voices with musical domain knowledge encoded in the fitness function. The limitations of using GA are that these systems are subjective and there is no way to simulate human behavior; furthermore, the user must hear all the potential solutions in order to evaluate a population. However, Biles claims that GA has efficient search method which is good for large scale searches and it has the ability to provide multiple solutions. (Bills, 2007).

Musical composition has been the focus of computer science since the 50s, and there exists several ways of representing music using computers. There have been several AI applications for automatic music compositions from early ages. When looking at AI applications, they can be categorized according to their most prominent features; namely, 1) Neural Networks (systems which learn), 2) Knowledge Based systems, 3) Evolutionary methods, and 4) Mathematical Models. Each of these models has their own pros and cons which were discussed in the above sections.

IIT Momentz

Good times good ppl :)

REQUIREMENTS BASED TESTING

IT project failure has become a very common factor during the past 10 years; statistics show that the root causes of many unsuccessful projects are due to poor requirements. James Martin shows in his study that 56% percent of bugs identified in projects are errors introduced in the requirements stage (Borland Software Corporation, 2005); therefore, getting the requirement right is essential. This can be achieved by requirements based testing approach, this essay is a discussion on what is requirement based testing, why it is important, how can it be applied to reduce bugs in software products, and what benefits can be gained.

 “Requirements Based Testing is the practice of aligning requirements at the beginning of a software project with quality expectations for testing –to ensure final deliverables consistently meet business needs.” (Alho, 2007). The above definition implies that using requirement based testing practices help ensure that the final outcome of the software consistently meets the business needs. Requirement based testing address two major concepts, first it focuses on validating whether the requirments are correct and makes sure that the requirements are up to the standards (ie complete,unambigous and logically consistent ), secondly it designs a set of test cases from the requirements. The design and the code should fully meet those requirements. This can be achived using black box testing. Therefore RBT can be defined as a methodology which defines set of rules to test the requirements and maximize the software quality by developing the right product.

However one can come to the conclution that the RBT process is tredious aditional work when developing a software product. But RBT is important because requirements are a critical factor when developing a software and you must get it right from the beginning of the development life cycle. Research has shown that most software failures are due to the following reasons

• Requirements and specifications are incomplete.
• Requirements and specifications change too often.
• There is a lack of user input (to requirements). (Mogyorodi, 2003)

All the above mention reasons are circled around requirements, therefore it is a critical need to verify and validate the requirements from the requirement stage itself.

Early bug detection reduces the cost of bug fixing; as mentioned above most root cause of bugs are due to inadequate requirements. According to the above definition of RBT, it improves the requirement and reduces software bugs from the requirement stage itself. This clarifies rather than following the traditional approach of testing which is performed after the code is delivered; therefore, it is important to follow the RBT process to eliminate software bugs and increase software quality.

To understand the RBT process it is important to know the RBT methodology. RBT methodology can be defined in 12 steps (Mogyorodi, 2003). Out of the 12 steps, five of the most important steps are taken in to consideration in-order to understand the effectiveness of the RBT process.

1.      Validation of requirements against business objectives – this ensures that the requirements satisfies the objectives of the project and meets the business need of the project.

2.      Review of requirements by stakeholders- refines the requirements by getting feedbacks from the stakeholder before going to the next phase.

3.      Development of use-cases - Map requirements against a task-oriented or interaction-oriented view of the system to make sure the requirements are complete.

4.      Application of language analysis techniques- Identify ambiguous, unclear or inconsistent phrases in the requirements and fix them to guarantee the consistency and clarity of the requirements.

5.      Logical consistency checks performed and test cases designed. - Design test cases that are 100% equivalent to the functionality in the requirements.

(Mogyorodi, 2003)

This gives a clear understanding on how the RBT process can improve the software quality and identify errors from the very first stage itself, due to validation of requirements against business objectives. This also indicates that the RBT process does not only detect bugs but it also address bug prevention by integrating testing throughout the development life cycle. At the same time RBT practices eliminates the most common risk of incomplete and changing requirement specifications issue because of its application of language analysis techniques and traceability functionalities. 

 

Figure 1: Requirements Based Testing process flow  (Aharonovitz, 2008)

Figure 1 show the actives which are undertaken by the RBT process. This shows the goals of RBT using separate boxes. Arrow heads direct how they can be achieved.

The first goal of the RBT process is ensuring the quality of requirement. It can be achieved by ensuring that business needs are met, maximizing user/domain expert’s involvement, and validating whether the requirements are correct and none conflicting. The following methods must be applied to ensure the quality of the requirements: Review of requirements against 

business objectives, by domain experts/business users, and map the requirements against use cases ambiguity analysis of requirements.  (Aharonovitz, 2008)

Designing logical test cases. This can be achieved by expressing the requirements using cause-effect cards, generation of test cases, and the use of truth tables or unique path detention can derive the equivalent test cases for the requirements to optimize the number of test cases.  (Aharonovitz, 2008)

The next two goals, ensuring the quality of the test cases, and the quality of design and code can be achieved by reviewing the test cases by requirement authors, domain experts, end users and also by reviewing test cases by designers and developers in the code review process.

Last goal of the RBT process is completing and executing tests which ensure that the actual behavior of the system is the expected behavior and meets the business needs. (Borland Software Corporation., 2006)

Moty Aharonovitz, senior director of Product Strategy at Borland Software, states that traceability plays a critical role in the RBT process; he claims that maintaining traceability information between requirements, test-cases and tests is crucial. This determines which test cases or tests should be changed when a specific requirement changes, and the impact of change. (Aharonovitz, 2008)

In conclusion RBT process and activities describe that RBT depends on two major functionalities 1) Quality assured requirements and 2) test cases. Using the following functionalities throughout the development life cycle development teams can guarantee that they are building the right product with the lowest defects.

To ease the RBT process and use it effectively organizations can use tools which automate the RBT process. For example Borland software cooperation has taken several approaches to automate the RBT process and has developed a set of tools, namely Borland Caliber DefineIT - designed to ensure that software requirements are defined completely and accurately from the beginning, Borland CaliberRM - designed to facilitate collaboration, impact analysis, and communication in the definition and management of changing requirements, and  SilkCentral Test Manager - delivers visibility and control of project quality aligned with business requirements and is integrated with CaliberRM to maintain bi-directional traceability and change impact analysis between business requirements and test activities (Borland Software

Corporation., 2006). By using an automated RBT process it would ease the software development cycle, and the final outcome will avoid downstream defects and project delays.

From the above discussion it is a fact that when a company follows RBT methodology, they will be rewarded with better products and profit. Considering the facts discussed above requirement based testing can be dived in to three forms from the business point of view; 1) Business benefits, 2) Operational benefits and 3) financial benefits.

According to the RBT methodology as discussed above each of the categorized benefits are analyzed using the RBT functionalities. Beginning with the business benefits which can be achieved by clear, complete and correct requirements which lead to developing what the customer really wants; thus resulting in higher customer satisfaction and improved market agility. Operational benefits focus on benefits while developing the software. As discussed above RBT has a well structured methodology to improve tractability and visibility of the requirements which eases the development cycle. This will optimize the QA activities ensuring the quality of the product. Furthermore it enhances the software quality by introducing requirements test coverage which will ensure all requirements are satisfied by the end product. Furthermore financial benefits can be looked at from the revenue prospect of an organization. By following the RBT process and activities development teams can reduce compliance risks and improve impact analysis. Moreover the most important aspect of financial benefits are related to profit;  by identifying the defects early and fixing them in a very early stages it reduces re-work, helping save time and money which will leads to gaining more profit.

Although there are many advantages of using the RBT process in the development life cycle Robin F. Goldsmith, JD claims that there can be draw backs associated with RBT process as well. He states that there are limitations imposed by inadequately defined requirements, assessing the extent to which the requirements are complete, the appropriate level of test case details, and inclusion of requirements based unit tests by developers. However he also states the above pitfalls can be overcome by identifying the strengths and often unrecognized weakness of requirement base tests, and understanding the importance of testing based on business as well as system requirements and discovering how to identify more of the necessary but often overlooked test cases (Goldsmith, 2008).

In conclusion the requirements based testing approach is a very important process to be followed due to the fact that many IT project failures are associated with requirements and requirement management. The requirements based testing process introduces a well formed procedure from the requirement gathering stage to the deploying stage of the product. It has an effective methodology to gather requirements without flaws, and defines how to test the requirements properly in-order to ensure that the right requirements are captured. At the same time it has methods to keep track of changing requirements, and ways to define requirement based test cases including maximum coverage with a minimum number of test cases. It is a fact that by following the above mentioned process most of the software bugs can be identified early, requirements specifications can be produced clearly, and requirements changes can be managed effectively. Even though there can be pit falls in the RBT process, if it is followed properly one can avoid many draw backs and gain a lot of benefits from it.