Spaces:
Sleeping
Sleeping
Raivis Dejus
commited on
Commit
•
622eda7
1
Parent(s):
306d4b2
Adjusting app
Browse files
app.py
CHANGED
@@ -98,14 +98,6 @@ demo = gr.Blocks()
|
|
98 |
transcribe = gr.Interface(
|
99 |
fn=transcribe,
|
100 |
inputs=[
|
101 |
-
# gr.Markdown(
|
102 |
-
# """
|
103 |
-
# Test Latvian speech recognition (STT) models. Three models are available:
|
104 |
-
# * [tiny](https://huggingface.co/RaivisDejus/whisper-tiny-lv) - Fastest, requiring least RAM, but also least accurate
|
105 |
-
# * [small](https://huggingface.co/RaivisDejus/whisper-small-lv) - Reasonably fast, reasonably accurate, requiring reasonable amounts of RAM
|
106 |
-
# * [large](https://huggingface.co/AiLab-IMCS-UL/whisper-large-v3-lv-late-cv17) - Most accurate, developed by scientists from [ailab.lv](https://ailab.lv/). Requires most RAM and for best performance should be run on a GPU.
|
107 |
-
# """
|
108 |
-
# ),
|
109 |
gr.Dropdown([
|
110 |
("tiny", "RaivisDejus/whisper-tiny-lv"),
|
111 |
("small", "RaivisDejus/whisper-small-lv"),
|
@@ -114,7 +106,7 @@ transcribe = gr.Interface(
|
|
114 |
gr.Audio(sources=["upload", "microphone"],type="filepath", label="Audio"),
|
115 |
gr.Radio([("Transcribe", "transcribe"), ("Translate to English", "translate",)], label="Task", value="transcribe"),
|
116 |
],
|
117 |
-
outputs=gr.Textbox(label="Transcription", lines=
|
118 |
title="Latvian speech recognition: Transcribe Audio",
|
119 |
description=("""
|
120 |
Test Latvian speech recognition (STT) models. Three models are available:
|
@@ -124,6 +116,8 @@ transcribe = gr.Interface(
|
|
124 |
* [small](https://huggingface.co/RaivisDejus/whisper-small-lv) - Reasonably fast, reasonably accurate, requiring reasonable amounts of RAM
|
125 |
|
126 |
* [large](https://huggingface.co/AiLab-IMCS-UL/whisper-large-v3-lv-late-cv17) - Most accurate, developed by scientists from [ailab.lv](https://ailab.lv/). Requires most RAM and for best performance should be run on a GPU.
|
|
|
|
|
127 |
"""
|
128 |
),
|
129 |
allow_flagging="never",
|
@@ -142,8 +136,17 @@ yt_transcribe = gr.Interface(
|
|
142 |
],
|
143 |
outputs=["html", "text"],
|
144 |
title="Latvian speech recognition: Transcribe YouTube",
|
145 |
-
description=(
|
146 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
147 |
),
|
148 |
allow_flagging="never",
|
149 |
)
|
|
|
98 |
transcribe = gr.Interface(
|
99 |
fn=transcribe,
|
100 |
inputs=[
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
101 |
gr.Dropdown([
|
102 |
("tiny", "RaivisDejus/whisper-tiny-lv"),
|
103 |
("small", "RaivisDejus/whisper-small-lv"),
|
|
|
106 |
gr.Audio(sources=["upload", "microphone"],type="filepath", label="Audio"),
|
107 |
gr.Radio([("Transcribe", "transcribe"), ("Translate to English", "translate",)], label="Task", value="transcribe"),
|
108 |
],
|
109 |
+
outputs=gr.Textbox(label="Transcription", lines=15),
|
110 |
title="Latvian speech recognition: Transcribe Audio",
|
111 |
description=("""
|
112 |
Test Latvian speech recognition (STT) models. Three models are available:
|
|
|
116 |
* [small](https://huggingface.co/RaivisDejus/whisper-small-lv) - Reasonably fast, reasonably accurate, requiring reasonable amounts of RAM
|
117 |
|
118 |
* [large](https://huggingface.co/AiLab-IMCS-UL/whisper-large-v3-lv-late-cv17) - Most accurate, developed by scientists from [ailab.lv](https://ailab.lv/). Requires most RAM and for best performance should be run on a GPU.
|
119 |
+
|
120 |
+
To improve speech recognition quality, more data is needed, donate your voice on [Balsu talka](https://balsutalka.lv/)
|
121 |
"""
|
122 |
),
|
123 |
allow_flagging="never",
|
|
|
136 |
],
|
137 |
outputs=["html", "text"],
|
138 |
title="Latvian speech recognition: Transcribe YouTube",
|
139 |
+
description=("""
|
140 |
+
Test Latvian speech recognition (STT) models. Three models are available:
|
141 |
+
|
142 |
+
* [tiny](https://huggingface.co/RaivisDejus/whisper-tiny-lv) - Fastest, requiring least RAM, but also least accurate
|
143 |
+
|
144 |
+
* [small](https://huggingface.co/RaivisDejus/whisper-small-lv) - Reasonably fast, reasonably accurate, requiring reasonable amounts of RAM
|
145 |
+
|
146 |
+
* [large](https://huggingface.co/AiLab-IMCS-UL/whisper-large-v3-lv-late-cv17) - Most accurate, developed by scientists from [ailab.lv](https://ailab.lv/). Requires most RAM and for best performance should be run on a GPU.
|
147 |
+
|
148 |
+
To improve speech recognition quality, more data is needed, donate your voice on [Balsu talka](https://balsutalka.lv/)
|
149 |
+
"""
|
150 |
),
|
151 |
allow_flagging="never",
|
152 |
)
|