Speaker App

The Speaker App lets you generate speech in many languages (more than 30 languages plus variations) and with selected voices, including SELMA voices (Portuguese for Brazil).

Screenshot of the SELMA Speaker App

Image shows the Speaker APP UI. On the left-hand side is a line of text (“Olá, meu nome é Leila.”). The text is generated using a SELMA voice (Leila). The audio can be listened to or downloaded as a WAV file.

SELMA contribution

  • Development of App UI
  • Integration of an API from SELMA OSS and other providers
  • User Evaluation for improved TTS (Text to Speech)

More

  • SELMA Speaker App (available on Apple TestFlight, credentials needed)

Bring me to Diarization Prototype