I worked out an example where i creating a speaker embedding for my own voice. I recorded some WAV and it kind of worked. However, while using that embedding, the output is like a pure robot. Am i missing something? should there be a particular speech?
Related topics
Topic | Replies | Views | Activity | |
---|---|---|---|---|
Help for using whisper with embeddings | 1 | 425 | November 22, 2023 | |
What are the parameters the pyannote embedding model was trained on? | 0 | 522 | August 6, 2023 | |
How to extract embeddings in Wav2Vec2? | 0 | 434 | April 29, 2022 | |
Can you use the Same embeddings of Wav2Vec XLSR and apply different ASR heads? | 0 | 241 | June 2, 2022 | |
WavLM ECAPA-TDNN embeddings for Speaker verification | 0 | 591 | November 19, 2023 |