Update architecture image (#13)

- Update architecture image (4980e89919e262e12442b4152164fe4f048db087)

Co-authored-by: Kenneth Hamilton <[email protected]>

Files changed (1) hide show

docs/architecture.md CHANGED Viewed

@@ -48,7 +48,7 @@ The AR model follows a Mistral-style decoder-only transformer model to predict E
 Overall, the AR and NAR model is going to predict all 8 codebook entries of the Encodec 6kbps codec.
 The AR model design is given below:
-![Mars 5 AR architecture](/docs/assets/mars5_AR_arch.png)
 **Figure**: autoregressive component of Mars 5. During training, the initial 6kbps encodec tokens of the speech are fed through a small encoder-only transformer, producing a single output vector corresponding to an implicit speaker embedding.
 This vector is concatenated with learnt embeddings corresponding to the text tokens, and L0 speech tokens, after byte-pair encoding tokenization.

 Overall, the AR and NAR model is going to predict all 8 codebook entries of the Encodec 6kbps codec.
 The AR model design is given below:
+![Mars 5 AR architecture](https://raw.githubusercontent.com/Camb-ai/MARS5-TTS/master/docs/assets/mars5_AR_arch.png)
 **Figure**: autoregressive component of Mars 5. During training, the initial 6kbps encodec tokens of the speech are fed through a small encoder-only transformer, producing a single output vector corresponding to an implicit speaker embedding.
 This vector is concatenated with learnt embeddings corresponding to the text tokens, and L0 speech tokens, after byte-pair encoding tokenization.