config, benchmark notebook, synthesis fixed

config.json

@@ -75,6 +75,7 @@
     "phoneme_cache_path": "mozilla_us_phonemes",  // phoneme computation is slow, therefore, it caches results in the given folder.
     "use_phonemes": true,           // use phonemes instead of raw characters. It is suggested for better pronounciation.
     "phoneme_language": "en-us",     // depending on your target language, pick one from  https://github.com/bootphon/phonemizer#languages
-    "text_cleaner": "phoneme_cleaners"
+    "text_cleaner": "phoneme_cleaners",
+    "num_speakers": 10 // should just be bigger than the actual number of speakers
 }
 

config_tacotron.json

@@ -76,6 +76,7 @@
         "phoneme_cache_path": "mozilla_us_phonemes",  // phoneme computation is slow, therefore, it caches results in the given folder.
         "use_phonemes": true,           // use phonemes instead of raw characters. It is suggested for better pronounciation.
         "phoneme_language": "en-us",     // depending on your target language, pick one from  https://github.com/bootphon/phonemizer#languages
-        "text_cleaner": "phoneme_cleaners"
+        "text_cleaner": "phoneme_cleaners",
+        "num_speakers": 10 // should just be bigger than the actual number of speakers
     }
     

config_tacotron2.json

@@ -78,6 +78,7 @@
         "phoneme_cache_path": "mozilla_us_phonemes",  // phoneme computation is slow, therefore, it caches results in the given folder.
         "use_phonemes": true,           // use phonemes instead of raw characters. It is suggested for better pronounciation.
         "phoneme_language": "en-us",     // depending on your target language, pick one from  https://github.com/bootphon/phonemizer#languages
-        "text_cleaner": "phoneme_cleaners"
+        "text_cleaner": "phoneme_cleaners",
+        "num_speakers": 10 // should just be bigger than the actual number of speakers
     }
     

config_tacotron_de.json

@@ -76,6 +76,7 @@
                 "phoneme_cache_path": "phoneme_cache",  // phoneme computation is slow, therefore, it caches results in the given folder.
                 "use_phonemes": false,           // use phonemes instead of raw characters. It is suggested for better pronounciation.
                 "phoneme_language": "de",     // depending on your target language, pick one from  https://github.com/bootphon/phonemizer#languages
-                "text_cleaner": "phoneme_cleaners"
+                "text_cleaner": "phoneme_cleaners",
+                "num_speakers": 10 // should just be bigger than the actual number of speakers
             }
             

config_tacotron_gst.json

@@ -76,6 +76,7 @@
         "phoneme_cache_path": "mozilla_us_phonemes",  // phoneme computation is slow, therefore, it caches results in the given folder.
         "use_phonemes": true,           // use phonemes instead of raw characters. It is suggested for better pronounciation.
         "phoneme_language": "en-us",     // depending on your target language, pick one from  https://github.com/bootphon/phonemizer#languages
-        "text_cleaner": "phoneme_cleaners"
+        "text_cleaner": "phoneme_cleaners",
+        "num_speakers": 10 // should just be bigger than the actual number of speakers
     }
     

utils/synthesis.py

@@ -35,17 +35,17 @@ def compute_style_mel(style_wav, ap, use_cuda):
         return style_mel
 
 
-def run_model(model, inputs, CONFIG, truncated, style_mel=None):
+def run_model(model, inputs, speaker_id, CONFIG, truncated, style_mel=None):
     if CONFIG.model == "TacotronGST" and style_mel is not None:
         decoder_output, postnet_output, alignments, stop_tokens = model.inference(
-            inputs, style_mel)
+            inputs, style_mel, speaker_id)
     else:
         if truncated:
             decoder_output, postnet_output, alignments, stop_tokens = model.inference_truncated(
-                inputs)
+                inputs, speaker_id)
         else:
             decoder_output, postnet_output, alignments, stop_tokens = model.inference(
-                inputs)
+                inputs, speaker_id)
     return decoder_output, postnet_output, alignments, stop_tokens
 
 
@@ -100,12 +100,13 @@ def synthesis(model,
         style_mel = compute_style_mel(style_wav, ap, use_cuda)
     # preprocess the given text
     inputs = text_to_seqvec(text, CONFIG, use_cuda)
-    speaker_id = speaker_id_var = torch.from_numpy(speaker_id).unsqueeze(0)
+    speaker_id = np.asarray(speaker_id)
+    speaker_id = torch.from_numpy(speaker_id).unsqueeze(0)
     if use_cuda:
         speaker_id.cuda()
     # synthesize voice
     decoder_output, postnet_output, alignments, stop_tokens = run_model(
-        model, inputs, CONFIG, truncated, style_mel)
+        model, inputs, speaker_id, CONFIG, truncated, style_mel)
     # convert outputs to numpy
     postnet_output, decoder_output, alignment = parse_outputs(
         postnet_output, decoder_output, alignments)