Fiund / Speech & conversation data
TTS training data
Scope the voice, recording conditions and intended use.
A clear review process
What to establish for your project.
Define the synthesis use case
Explain whether the project concerns a general speech model, a named voice, adaptation or evaluation. These uses can require different source choices and permission scopes. Do not assume that permission to use a podcast as listening material covers a synthesis use case.
Describe the source signal
Set requirements for isolated speakers, room conditions, noise, encoding, sample rate and delivery format. Clearly distinguish original participant tracks from processed mixtures and derived segments.
Plan text and phonetic coverage
Discuss transcript accuracy, pronunciation variation, numbers, names, punctuation and the relationship between text and the recorded utterance. Request coverage evidence for the features your model needs rather than relying on a total-hour figure.
Review the selected material
Assess representative recordings against the actual synthesis pipeline. Establish what source and participant documentation is available, the exact permitted use and any restrictions in the project agreement.
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.