Dataset focus
ASR training data
Conversational speech for recognition models.
Podcast audio · Speech models · Conversational AI
Podcast audio.
Built around
your model.
Define the conversations, speakers, languages and formats your team needs. Work directly with Fiund on source fit and a documented licensing scope.
AIPodcast is a specialist Fiund website for podcast audio and conversational speech sourcing.
Source recordings
Discuss sessions, episodes and participant tracks.
Technical detail
Specify channels, transcripts and speaker labels.
Reviewable scope
Connect selected material with the intended use.
Direct contact
Speak with Jaeden Schafer at Fiund.
Dataset sourcing
Start with the model task.
Dataset focus
TTS training data
Scope the voice, recording conditions and intended use.
Dataset focus
Conversational AI datasets
Preserve the interaction, not just individual utterances.
Dataset focus
Multilingual speech data
Define meaningful language and regional coverage.
Dataset focus
Podcast audio licensing
Bring source recordings and intended use into the same brief.
Dataset focus
Speech data for AI
A specification that connects the source to the model task.
Founder, Fiund
Your project contact
For research, engineering & procurement teams
A direct conversation.
A well-defined next step.
Speak directly with Jaeden Schafer about the recordings, documentation and preparation your speech-data project needs. Start with the model task and agree on a clear source-review and commercial scope.
Technical requirementsDefine coverage, formats, annotations and acceptance criteria.
Source & permission reviewEstablish the documentation needed for your intended use.
Commercial scopeDiscuss samples, preparation, delivery and project-specific terms.
Collection design
The right recording level
changes the project.
Source level
Published episodes
Specify editorial content, music, advertising, compression and exclusions when considering a finished episode.
Source level
Original recording sessions
Discuss unique sessions, speaker relationships, timestamps and any alternate edits of the same conversation.
Source level
Participant tracks
State whether independent speaker audio is essential. It should be demonstrated on the proposed source, rather than inferred from the show format.
Original recording
A real voice. A visible source.
From Jaeden’s recording archive
Hear a real spoken-word excerpt.
A short AI Chat excerpt recorded by Jaeden Schafer. This is an editorial example of original speech; a requested dataset has its own source and permission review.
Read the automatic transcript
Today on the podcast, I wanted to do a big update on xAI, some things happening over at X, as I think this is a pretty significant amount of money that has just been put into the ecosystem.
Working with Fiund
A clear route from brief
to a defined proposal.
01 / Define
Write the source brief
Specify the model task, languages, recording level, unique speakers, technical formats and intended use.
02 / Review
Assess the proposed material
Discuss representative review, source relationships, participant documentation and exclusions relevant to the collection.
03 / Agree
Confirm the project scope
Establish preparation, acceptance criteria, commercial terms and delivery arrangements for the selected source.
Start a project
What should a useful sample demonstrate?
Tell us the signal, source characteristics and review criteria your team needs to inspect. We can then discuss an appropriate next step.
Resource library
Make the first conversation concrete.
Field guide
Understand ASR training data
Define the relationship between recordings, text and the recognition task.
Field guide
Review speech dataset quality
Consider coverage, signal quality, annotations and source relationships.
Field guide
Prepare a licensing conversation
Bring intended uses, source documentation and commercial requirements into the brief.