Spaces:
Running
Running
Update app.py (#10)
Browse files
app.py
CHANGED
@@ -125,7 +125,7 @@ router = Client("ttseval/tts-router", hf_token=os.getenv('HF_TOKEN'))
|
|
125 |
####################################
|
126 |
MUST_BE_LOGGEDIN = "Please login with Hugging Face to participate in the TTS Arena."
|
127 |
DESCR = """
|
128 |
-
# TTS Arena: Benchmarking TTS models in the Wild
|
129 |
|
130 |
Vote to find the best Text-to-Speech model out there!
|
131 |
""".strip()
|
@@ -141,13 +141,12 @@ Vote to find the best Text-to-Speech model out there!
|
|
141 |
INSTR = """
|
142 |
## Instructions
|
143 |
|
144 |
-
* Input the text to
|
145 |
* Listen to the two audio clips, one after the other.
|
146 |
* Vote on which audio sounds more natural to you.
|
|
|
147 |
|
148 |
-
|
149 |
-
|
150 |
-
## Synthesize now!
|
151 |
""".strip()
|
152 |
request = ''
|
153 |
if SPACE_ID:
|
@@ -157,7 +156,7 @@ if SPACE_ID:
|
|
157 |
Please fill out [this form](https://huggingface.co/spaces/{SPACE_ID}/discussions/new?title=%5BModel+Request%5D+&description=%23%23%20Model%20Request%0A%0A%2A%2AModel%20website%2Fpaper%20%28if%20applicable%29%2A%2A%3A%0A%2A%2AModel%20available%20on%2A%2A%3A%20%28coqui%7CHF%20pipeline%7Ccustom%20code%29%0A%2A%2AWhy%20do%20you%20want%20this%20model%20added%3F%2A%2A%0A%2A%2AComments%3A%2A%2A) to request a model.
|
158 |
"""
|
159 |
ABOUT = f"""
|
160 |
-
## About
|
161 |
|
162 |
The TTS Arena is a project created to evaluate leading speech synthesis models. It is inspired by the [Chatbot Arena](https://chat.lmsys.org/) by LMSys.
|
163 |
|
@@ -187,11 +186,11 @@ We may store text you enter and generated audio. We store a unique ID for each s
|
|
187 |
Generated audio clips cannot be redistributed and used for personal, non-commercial use only.
|
188 |
""".strip()
|
189 |
LDESC = """
|
190 |
-
## Leaderboard
|
191 |
|
192 |
-
A list of the models
|
193 |
|
194 |
-
|
195 |
|
196 |
Tick the `Reveal Preliminary Results` checkbox below if you wish to see the raw data.
|
197 |
""".strip()
|
|
|
125 |
####################################
|
126 |
MUST_BE_LOGGEDIN = "Please login with Hugging Face to participate in the TTS Arena."
|
127 |
DESCR = """
|
128 |
+
# TTS Arena: Benchmarking TTS models in the Wild β‘
|
129 |
|
130 |
Vote to find the best Text-to-Speech model out there!
|
131 |
""".strip()
|
|
|
141 |
INSTR = """
|
142 |
## Instructions
|
143 |
|
144 |
+
* Input the text to synthesise audio (or press π² for a random text).
|
145 |
* Listen to the two audio clips, one after the other.
|
146 |
* Vote on which audio sounds more natural to you.
|
147 |
+
* Model names are revealed after the vote is cast.
|
148 |
|
149 |
+
## Synthesise now!
|
|
|
|
|
150 |
""".strip()
|
151 |
request = ''
|
152 |
if SPACE_ID:
|
|
|
156 |
Please fill out [this form](https://huggingface.co/spaces/{SPACE_ID}/discussions/new?title=%5BModel+Request%5D+&description=%23%23%20Model%20Request%0A%0A%2A%2AModel%20website%2Fpaper%20%28if%20applicable%29%2A%2A%3A%0A%2A%2AModel%20available%20on%2A%2A%3A%20%28coqui%7CHF%20pipeline%7Ccustom%20code%29%0A%2A%2AWhy%20do%20you%20want%20this%20model%20added%3F%2A%2A%0A%2A%2AComments%3A%2A%2A) to request a model.
|
157 |
"""
|
158 |
ABOUT = f"""
|
159 |
+
## π About
|
160 |
|
161 |
The TTS Arena is a project created to evaluate leading speech synthesis models. It is inspired by the [Chatbot Arena](https://chat.lmsys.org/) by LMSys.
|
162 |
|
|
|
186 |
Generated audio clips cannot be redistributed and used for personal, non-commercial use only.
|
187 |
""".strip()
|
188 |
LDESC = """
|
189 |
+
## ποΈ Leaderboard
|
190 |
|
191 |
+
A list of the models based on how natural they sound (according to the votes cast)!
|
192 |
|
193 |
+
Important: The leaderboard will be hidden by default until a human ratings threshold has been achieved to keep the results fair.
|
194 |
|
195 |
Tick the `Reveal Preliminary Results` checkbox below if you wish to see the raw data.
|
196 |
""".strip()
|