T3 Talk2Text - A model for near real-time voice transcription in virtual group meetings

Group projects thrive on communication, but how can students revisit discussions effortlessly? We present T3 Talk2Text, an open-source tool for real-time meeting transcription, enabling reflection and collaboration analysis. Discover the tech behind it!
Like

Share this post

Choose a social network to share with, or copy the URL to share elsewhere

This is a representation of how your post may appear on social media. The actual post will vary between social networks

Explore the Research

SpringerLink
SpringerLink SpringerLink

T3 Talk2Text – A model for near real-time voice transcription in virtual group meetings - Discover Education

Group projects are an important part of many educational programs. In these projects, groupware tools like shared workspaces, shared editors, and synchronous or asynchronous communication tools (video conferencing, chat, email, forum) are often utilized. For synchronous collaboration, video conferencing systems are often used since they allow for direct and effortless informal communication via speech and video. If the group is to reflect about the content and way of communication and collaboration within and across meetings, the group needs access to the development of artifacts as well as the conversation of the meetings. While some video conferencing systems allow recording of a meeting, accessing content and dialogue structure requires real-time viewing and detailed note-taking. To reduce this effort, a video chat should include a transcription functionality that is able to support students in an online group work session by providing a video chat with automatic creation of a transcript. For this purpose, a video communication tool named T3 was developed that enables communication between learners and generates a transcript through an integrated automatic speech recognition (ASR) system. The transcript can be used to reflect on the conversations in order to recall the topics discussed or to identify group work problems, such as insufficient participation, coordination and collaboration problems. The implementation of T3 uses voice activation detection, WebRTC and ASR models to maintain a high level of quality in the transcription process. Initial functionality tests demonstrated the ability of T3 to create accurate group discussion transcripts, making it attractive for students and teachers to assess and improve communication and collaboration, and for researchers studying group discussions.

The Challenge: Capturing Group Discussions for Reflection

Group projects are a cornerstone of modern education, but remote collaboration introduces challenges like unequal participation, miscommunication, and the difficulty of recalling past discussions. While video conferencing tools facilitate real-time interaction, they lack built-in support for documenting conversations. Manual note-taking is cumbersome, and post-meeting transcriptions from recordings are time-consuming.

We asked: How can we provide students and educators with an effortless way to capture and reflect on group discussions in near real-time?

Introducing T3 Talk2Text

t3 - logo

Our solution, T3 Talk2Text, is an open-source web application that integrates:

  • WebRTC-based video conferencing (peer-to-peer, no expensive licenses)
  • Automatic Speech Recognition (ASR) and Real-time transcription via OpenAI’s Whisper model
  • On-demand summaries (using Llama3 for AI-generated insights)

Unlike commercial tools (e.g., Microsoft Teams), T3 prioritizes privacy (self-hosted), multilingual support, and accessibility (works on any device with a browser).

 User interface of T3 - T3 in use                      

Key Innovations

  1. Voice Activity Detection (VAD) Pipeline

    • Filters background noise and segments speech for accurate ASR input.

    • Achieves 8.04% Word Error Rate (WER) in German—comparable to human transcribers.

  2. Dynamic Transcript Formats
    Users can download transcripts as:

    • PDFs (messenger-style, with speaker alignment)

    • CSVs (structured for analysis)

    • AI summaries (condensed key points)

      Possible dialog protocol output formats

  3. Scalable Architecture

    • Lightweight SQLite storage for transcripts.

    • Peer-to-peer media streaming reduces server load.

Behind the Scenes: Overcoming Technical Hurdles

Challenge 1: Real-Time Processing
WebRTC’s low-latency streams were ideal for video chat but required careful buffering to feed Whisper ASR without delays. Our VAD component ensured only speech segments were processed, optimizing resource use.

Challenge 2: Multilingual Support
Whisper’s multilingual capabilities let T3 adapt to diverse classrooms, though future work will explore fine-tuning for non-native accents.

Challenge 3: Privacy-First Design
All data stays on institutional servers, and temporary audio files are deleted post-transcription, which is critical for GDPR compliance.

Impact and Future Directions

Initial tests with student groups showed T3 seamlessly integrated into discussions without disrupting collaboration. Educators highlighted its potential for:

  • Identifying participation gaps (via speaker-labeled transcripts).

  • Conflict resolution (revisiting past dialogue).

  • Research (analyzing communication patterns across courses).

Next steps include:

  • Deploying Whisper Large Turbo for faster, even more accurate transcriptions.

  • Longitudinal studies in university courses to measure learning outcomes.

Try It Out!

T3 is open-source and available for institutions to adapt. We welcome collaborations to explore its use in classrooms and beyond.

Read the full paper: The Paper
Code repository: Github
Contact: Thomas.Kasakowskij@fernuni-hagen.de

Follow the Topic

Speech and Audio Processing
Technology and Engineering > Electrical and Electronic Engineering > Signal, Speech and Image Processing > Speech and Audio Processing
eLearning
Humanities and Social Sciences > Education > Media Education > Digital Education and Educational Technology > eLearning
Groupwork and Presentation
Humanities and Social Sciences > Education > Skills > Groupwork and Presentation
Higher Education
Humanities and Social Sciences > Education > Higher Education
User Interfaces and Human Computer Interaction
Mathematics and Computing > Computer Science > Computer and Information Systems Applications > User Interfaces and Human Computer Interaction
Natural Language Processing (NLP)
Mathematics and Computing > Computer Science > Artificial Intelligence > Natural Language Processing (NLP)

Related Collections

With Collections, you can get published faster and increase your visibility.

Advancing Teaching & Learning: Integrating AI, Emerging Technologies, and Pedagogical Innovation

Collection "Advancing Teaching & Learning: AI, Emerging Technologies and Pedagogical Innovation" addresses a major challenge in education today. There is a lack of resources that bring advanced technologies and innovative teaching together. AI, digital tools and emerging technologies are transforming classrooms. Yet many teachers and education leaders struggle to use these tools effectively. This Collection fills that gap. It offers expert insights, practical strategies and real-world examples for modern teaching and learning.

The Collection is linked to the United Nations Sustainable Development Goal 4 on quality education. It shows how to make learning more accessible, inclusive and effective for all. AI, adaptive learning platforms and immersive technologies can personalise instruction. They can engage students and support diverse needs. At the same time, the human element of teaching remains central.

Readers will find practical guidance on curriculum design, digital literacy and evidence-based teaching methods. The approaches can be applied in classroom, blended and online learning. Beyond tools, the Collection inspires educators to rethink traditional methods. It encourages them to embrace innovation while keeping learning meaningful and student-focused.

By addressing the lack of comprehensive resources on advanced technologies, Advancing Teaching & Learning helps teachers, school leaders and instructional designers navigate the digital age. It equips them to create engaging, future-ready learning experiences. Students are prepared for a connected, digital world. Educators gain the confidence to unlock the full potential of every learner.

This Collection supports and amplifies research related to the United Nations Sustainable Development Goal 4 on quality education. It shows how to make learning more accessible, inclusive and effective for all. AI, adaptive learning platforms and immersive technologies can personalise instruction. They can engage students and support diverse needs. At the same time, the human element of teaching remains central.

Keywords: Teaching and Learning; AI in Education; Emerging Technologies; Digital Tools in Classrooms; Pedagogical Innovation; Adaptive Learning Platforms; Immersive Technologies; Quality Education (SDG 4); Classroom Innovation; Blended and Online Learning; Student Engagement; Teacher Professional Development; Future-Ready Learning; Instructional Design; Educational Technology Integration.

Publishing Model: Open Access

Deadline: Apr 15, 2027

Advancing Health Professions Education: Innovations, Strategies, and Global Perspectives

Health Professions Education (HPE) plays an important role in preparing competent, ethical, and compassionate healthcare professionals capable of responding to the complex demands of contemporary healthcare systems. In recent years, the field of HPE has witnessed a paradigm shift towards learner-centred, competency-based, and interprofessional education models that prioritize not only clinical expertise but also communication, professionalism, teamwork, and leadership skills.

This Collection invites original research, systematic or scoping reviews, case studies and brief reports and that critically examine the evolving landscape of HPE across medicine, dentistry, nursing, pharmacy, and other allied health disciplines. Key areas of interest include curriculum design and reform, innovative teaching and assessment methods, integration of digital and blended learning technologies, interprofessional education (IPE), faculty development, educational management, student support and well-being, and strategies to promote equity, diversity, and inclusivity in HPE.

We also welcome contributions that explore challenges and best practices in clinical training, rural and underserved community placements, global health education, policy development, accreditation, and institutional leadership in health education. Studies from low- and middle-income countries are strongly encouraged to enhance global representation and to promote scholarly exchange on scalable and contextually relevant innovations.

By consolidating scholarly work in this Collection, we aim to advance the discourse on high-quality, evidence-informed HPE that supports the achievement of Sustainable Development Goals (SDG 3 and SDG 4) and ultimately contributes to improved healthcare outcomes worldwide.

Publishing Model: Open Access

Deadline: Nov 30, 2026