During the period 2021—2024, the VR-funded project Gawarbati: Documenting a vulnerable linguistic community in the Hindu Kush (2020-01500) was carried out in collaboration between the Department of Linguistics (SU) and the Pakistan-based language resource center Forum for Language Initiatives with Henrik Liljegren as its principal investigator. This documentation project has laid a foundational framework for creating a lasting record of Gawarbati through the creation of an annotated audio and video corpus. However, the sheer volume of manual annotations required create an immense bottleneck for such field-linguistic work. In a pilot study the use of ASR to expedite this process by fine-tuning the Whisper-tiny model was carried out on 14 hours of Gawarbati data. The purpose of the current project, which primarily will be carried out within a master's thesis work, is to expand upon these initial findings by developing a multi-stage computational pipeline, as well as deliver a simple web application that can be utilized in future language documentation efforts for automatic annotation. Primary emphasis will be on ASR, and exploratory evaluation of automated IPA transcription or translation as annotation assistance tools.