Cofounder, Abridge
Search every verified Florian Metze interview, podcast appearance, and on-the-record quote — each transcript cross-checked by AI and human review to confirm speaker identity. Florian Metze, a cofounder of Abridge, discussed his career trajectory and current work in a September 2023 talk at Carnegie Mellon University. He recounted his early research in speech recognition with articulatory features, his later discovery of connectionist temporal classification, and his work on multimedia summarization and video retrieval. Metze stated that Carnegie Mellon provided him the opportunity to co-found a company focused on medical summarization, which he described as an effort to improve healthcare by making it easier to document doctor-patient conversations and summarize the information from those conversations. He also mentioned using his skills to recommend short-form videos to users at Meta in New York. In a separate 2016 presentation on visual features for context-aware speech recognition, Metze described research conducted with students on adapting speech recognizers using speaker, microphone distance, and visual features from videos. He discussed methods for training deep neural networks with additive feature shifts and using long short-term memory networks for adaptation, noting that the approach aimed to improve recognition of consumer-generated content like YouTube videos. He also suggested that visual context information could be particularly useful for low-resource languages.
Abstract Transcribing consumer generated multi-media content such as “Youtube” videos is still one of the hardest tasks for ...
Sign in to search the full transcript archive, filter by topic, and access every quote from Florian Metze.
The summary and quote tags on this profile are produced with AI assistance from verified, first-person interview transcripts, then checked by our team to confirm the speaker's identity and the accuracy of every quote. See how we verify →
Pro unlocks every transcript in full, plus unlimited search across every executive and company we cover.
50% off appliedPulling data at scale? The Data API gives you 30,000 transcripts a month over REST, MCP and JSON.
Compare all plans