From b14c11572ef96eb18c7c0ee7c3538f5d911caa8a Mon Sep 17 00:00:00 2001 From: Eren Golge Date: Tue, 14 May 2019 18:08:57 +0200 Subject: [PATCH] config update --- config.json | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/config.json b/config.json index 3c66beb4..b01ea0d5 100644 --- a/config.json +++ b/config.json @@ -1,6 +1,6 @@ { - "run_name": "mozilla-no-loc", - "run_description": "using Bahdenau attention, with original prenet.", + "run_name": "mozilla-no-loc-fattn-stopnet", + "run_description": "using forward attention, with original prenet, merged stopnet. Compare this with ", "audio":{ // Audio processing parameters @@ -42,12 +42,12 @@ "attention_norm": "softmax", // softmax or sigmoid. Suggested to use softmax for Tacotron2 and sigmoid for Tacotron. "prenet_type": "original", // ONLY TACOTRON2 - "original" or "bn". "prenet_dropout": true, // ONLY TACOTRON2 - enable/disable dropout at prenet. - "use_forward_attn": false, // ONLY TACOTRON2 - if it uses forward attention. In general, it aligns faster. + "use_forward_attn": true, // ONLY TACOTRON2 - if it uses forward attention. In general, it aligns faster. "transition_agent": false, // 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": false, // enable / disable loss masking against the sequence padding. "enable_eos_bos_chars": false, // enable/disable beginning of sentence and end of sentence chars. - "stopnet": false, // Train stopnet predicting the end of synthesis. + "stopnet": true, // Train stopnet predicting the end of synthesis. "separate_stopnet": false, // Train stopnet seperately if 'stopnet==true'. It prevents stopnet loss to influence the rest of the model. It causes a better model, but it trains SLOWER. "batch_size": 32, // Batch size for training. Lower values than 32 might cause hard to learn attention.