A database containing recordings of L2 English speakers producing the 720 ‘Harvard Sentences’. This project was created in the context of the seminar ‘Laboratory Phonology’ by Dr. Christopher Geissler at HHU University Düsseldorf - 2023 https://osf.io/u3ecd/
A dataset of L2 English speakers producing 70 sentences. This project was created in the context of the seminar ‘Applied Phonology: Evaluating Voice Assistants’ by Akhilesh Kakolu Ramarao M.A. and Erdin Mujezinović M.A. at HHU University Düsseldorf - 2025 (OSF link)
In this MaxMSP project users can listen to the audio of vowels from the vowel database provided by the UT Dallas and try to recreate them by changing the values of the formants F1, F2, F3 and F4. The values of the formants were extracted in Praat. The creators of the vowel database are Peter F. Assmann & William Katz. You can find it here: https://personal.utdallas.edu/~assmann/KIDVOW/index.html. The project is available on Github aswell as OSF. - 2025 https://github.com/andreoppenheimer/MaxMSP_Vowel_Synthesizer
A website built with Python, HTML and CSS via Flask that uses OPENAI-API to mimic a conversation with a desired streamer. This project was created in the context of the seminar ‘Informatik für Nicht-Informatiker (Programmierprojekte)’ by Dr. Carl Friedrich Bolz-Tereick at HHU University Düsseldorf. Users are able to select a streamer by typing their name. Furthermore they can download the conversation and upload it to continue where they left off. - 2025 https://github.com/andreoppenheimer/streamer_chatbot
The synthesizer allows for formants (F1, F2) and pitch to be manipulated independently. The build includes a task where synthesized reference vowels need to be recreated by the user. This project was created in the context of the seminar ‘Vocal Alchemy: Shaping Voices with Signal Processing’ by Prof. Dr. Kevin Tang, Prof. Dr. Florin Esther and Dr. Carter Williams at HHU University Düsseldorf. - 2025 https://osf.io/renqu/
A program written in Python designed to assisst in luinguistic field studies. The user can choose a word with a length between 1 to 6 syllables. The program then randomly combines linguistic features from German and English to genereate a IPA based transcription of a non-word of the desired length. The user can then apply multiple phonological rules. The results are all logged in a .txt file. One could use the produced non-word to assess through questioning L1 and L2 speakers of both languages which linguistic features hold more weight in assigning the word to one language or the other. This project was created in the context of the seminar ‘Python for Linguists’ by Prof. Dr. Kevin Tang at HHU University Düsseldorf. - 2025 https://osf.io/4dfhs/