Every interview you've recorded, in one searchable corpus
Phonotheca turns recordings into speaker-labeled transcripts you can code with tags, search across projects, and cite with timestamps. Built for qualitative research.
Interview 07 · First-generation students study54:12
12:04
Interviewer
You said the first semester felt like a different language.
12:11
P07
Everyone else seemed to have a manual I never got. I stopped asking questions in lecture and saved them for the library at night.
Belonging12:39
P07
My mother would call every Sunday and ask what I was learning. I translated it twice, once into Spanish, once into words that wouldn't worry her.
Family translation1
Upload a recording
Drop in the audio from your recorder or phone (mp3, m4a, wav, and the rest). Each project keeps a study's interviews together.
2
Transcription is included
Interviews transcribe with speaker labels in dozens of languages, processed in the EU and never retained after processing. Three audio hours are included every month.
3
Code, search, cite
Tag passages against your codebook, search the whole corpus, and export coded quotes with speaker and timestamp citations.
Lighter than a QDA suite, deeper than a transcription app
Transcription apps hand you one file at a time and meter every upload. QDA suites start at hundreds per license. Phonotheca is the middle: an archive that grows with your study, with the light coding a transcript needs.
Search the whole corpus
Full-text search across every transcript in every project. Find the passage you half-remember from interview four.
Tags are your codebook
Define themes once per project, in highlighter colors. Click a passage to code it; the mark stays visible as you read.
Quotes with citations
Copy any passage with speaker, interview, and timestamp attached, or export a project's coded quotes as Markdown or CSV.
Speaker-labeled transcripts
Diarization keeps interviewer and participant apart, and one click renames Speaker A across the transcript. The audio stays a click away for verification.
Hours, never file caps
Every limit is audio hours, not file counts. The archive and search are never gated: your corpus stays fully usable in months with zero new interviews.
Written for ethics review
EU-resident by design: audio is processed in-region and not retained after transcription. The data-handling statement is quotable in your data-management plan.