diff --git a/config_cluster.json b/config_cluster.json index d8b066d7..e651fe25 100644 --- a/config_cluster.json +++ b/config_cluster.json @@ -1,6 +1,6 @@ { - "run_name": "mozilla-nomask-fattn-bn", - "run_description": "Finetune 4702 orignal -> bn prenet - Mozilla with prenet bn, no mask, batch group size 0", + "run_name": "mozilla-fattn-agent-masking", + "run_description": "Original prenet,fattn and fattn agent with loss masking.", "audio":{ // Audio processing parameters @@ -40,10 +40,11 @@ "windowing": false, // Enables attention windowing. Used only in eval mode. "memory_size": 5, // ONLY TACOTRON - memory queue size used to queue network predictions to feed autoregressive connection. Useful if r < 5. "attention_norm": "softmax", // softmax or sigmoid. Suggested to use softmax for Tacotron2 and sigmoid for Tacotron. - "prenet_type": "bn", // ONLY TACOTRON2 - "original" or "bn". - "use_forward_attn": false, // ONLY TACOTRON2 - if it uses forward attention. In general, it aligns faster. - "transition_agent": false, // ONLY TACOTRON2 - enable/disable transition agent of forward attention. - "loss_masking": false, // enable / disable loss masking against the sequence padding. + "prenet_type": "original", // ONLY TACOTRON2 - "original" or "bn". + "use_forward_attn": true, // ONLY TACOTRON2 - if it uses forward attention. In general, it aligns faster. + "transition_agent": true, // ONLY TACOTRON2 - enable/disable transition agent of forward attention. + "location_attn": false, // ONLY TACOTRON2 - enable_disable location sensitive attention. It is enabled for TACOTRON by default. + "loss_masking": true, // enable / disable loss masking against the sequence padding. "enable_eos_bos_chars": false, // enable/disable beginning of sentence and end of sentence chars. "batch_size": 24, // Batch size for training. Lower values than 32 might cause hard to learn attention.