Merge pull request #611 from coqui-ai/dev

v0.1.1

Author: erogol

MANIFEST.in

@@ -1,6 +1,7 @@
 include README.md
 include LICENSE.txt
 include requirements.*.txt
+include requirements.txt
 include TTS/VERSION
 recursive-include TTS *.json
 recursive-include TTS *.html

Makefile

@@ -34,6 +34,9 @@ dev-deps:  ## install development deps
 doc-deps:  ## install docs dependencies
 	pip install -r docs/requirements.txt
 
+build-docs: ## build the docs
+	cd docs && make clean && make build
+
 hub-deps:  ## install deps for torch hub use
 	pip install -r requirements.hub.txt
 

TTS/.models.json

@@ -51,7 +51,7 @@
             "vctk":{
                 "sc-glow-tts": {
                     "description": "Multi-Speaker Transformers based SC-Glow model from https://arxiv.org/abs/2104.05557.",
-                    "github_rls_url": "https://github.com/coqui-ai/TTS/releases/download/v0.1.0/tts_models--en--vctk--sc-glowtts-transformer.zip",
+                    "github_rls_url": "https://github.com/coqui-ai/TTS/releases/download/v0.1.0/tts_models--en--vctk--sc-glow-tts.zip",
                     "default_vocoder": "vocoder_models/en/vctk/hifigan_v2",
                     "commit": "b531fa69",
                     "author": "Edresson Casanova",

TTS/VERSION

@@ -1 +1 @@
-0.1.0
+0.1.1

TTS/utils/trainer_utils.py

@@ -11,11 +11,15 @@ def is_apex_available():
 
 
 def setup_torch_training_env(cudnn_enable, cudnn_benchmark):
+    num_gpus = torch.cuda.device_count()
+    if num_gpus > 1:
+        raise RuntimeError(
+            f" [!] {num_gpus} active GPUs. Define the target GPU by `CUDA_VISIBLE_DEVICES`. For multi-gpu training use `TTS/bin/distribute.py`."
+        )
     torch.backends.cudnn.enabled = cudnn_enable
     torch.backends.cudnn.benchmark = cudnn_benchmark
     torch.manual_seed(54321)
     use_cuda = torch.cuda.is_available()
-    num_gpus = torch.cuda.device_count()
     print(" > Using CUDA: ", use_cuda)
     print(" > Number of GPUs: ", num_gpus)
     return use_cuda, num_gpus

tests/inference_tests/test_synthesize.py

@@ -10,19 +10,19 @@ def test_synthesize():
 
     # single speaker model
     run_cli(f'tts --text "This is an example." --out_path "{output_path}"')
-    # run_cli(
-    #     "tts --model_name tts_models/en/ljspeech/speedy-speech-wn "
-    #     f'--text "This is an example." --out_path "{output_path}"'
-    # )
-    # run_cli(
-    #     "tts --model_name tts_models/en/ljspeech/speedy-speech-wn  "
-    #     "--vocoder_name vocoder_models/en/ljspeech/multiband-melgan "
-    #     f'--text "This is an example." --out_path "{output_path}"'
-    # )
+    run_cli(
+        "tts --model_name tts_models/en/ljspeech/speedy-speech-wn "
+        f'--text "This is an example." --out_path "{output_path}"'
+    )
+    run_cli(
+        "tts --model_name tts_models/en/ljspeech/speedy-speech-wn  "
+        "--vocoder_name vocoder_models/en/ljspeech/multiband-melgan "
+        f'--text "This is an example." --out_path "{output_path}"'
+    )
 
-    # # multi-speaker model
-    # run_cli("tts --model_name tts_models/en/vctk/sc-glow-tts --list_speaker_idxs")
-    # run_cli(
-    #     f'tts --model_name tts_models/en/vctk/sc-glow-tts --speaker_idx "p304" '
-    #     f'--text "This is an example." --out_path "{output_path}"'
-    # )
+    # multi-speaker model
+    run_cli("tts --model_name tts_models/en/vctk/sc-glow-tts --list_speaker_idxs")
+    run_cli(
+        f'tts --model_name tts_models/en/vctk/sc-glow-tts --speaker_idx "p304" '
+        f'--text "This is an example." --out_path "{output_path}"'
+    )