Upload README.md
Browse files
README.md
CHANGED
@@ -23,7 +23,7 @@ pipeline_tag: text-to-speech
|
|
23 |
| Model | Published | Training Data | Compute (A100 80GB) | Released Voices | Released Langs |
|
24 |
| ----- | --------- | ------------- | ------------------- | --------------- | -------------- |
|
25 |
| v0.19 | 2024 Dec 25 | <100 hrs | 500 hrs @ $400 | 10 | 1 |
|
26 |
-
| **v1.0** | 2025 Jan 27 | Few hundred hrs | 1000 hrs @ $1000 | [
|
27 |
|
28 |
Training is continuous. The v0.19 model was produced "on the way" to the v1.0 model, so the Compute footprints overlap.
|
29 |
|
@@ -46,7 +46,7 @@ Most voices perform best on a "goldilocks range" of 100-200 tokens out of ~500 p
|
|
46 |
The following can be run in a single cell on [Google Colab](https://colab.research.google.com/).
|
47 |
```py
|
48 |
# 1️⃣ Install kokoro
|
49 |
-
!pip install -q kokoro soundfile
|
50 |
# 2️⃣ Install espeak, used for out-of-dictionary fallback
|
51 |
!apt-get -qq -y install espeak-ng > /dev/null 2>&1
|
52 |
# You can skip espeak installation, but OOD words will be skipped unless you provide a fallback
|
@@ -57,6 +57,8 @@ from IPython.display import display, Audio
|
|
57 |
import soundfile as sf
|
58 |
# 🇺🇸 'a' => American English
|
59 |
# 🇬🇧 'b' => British English
|
|
|
|
|
60 |
pipeline = KPipeline(lang_code='a') # make sure lang_code matches voice
|
61 |
|
62 |
# The following text is for demonstration purposes only, unseen during training
|
@@ -67,10 +69,12 @@ It was a Sprawl voice and a Sprawl joke. The Chatsubo was a bar for professional
|
|
67 |
|
68 |
These were to have an enormous impact, not only because they were associated with Constantine, but also because, as in so many other areas, the decisions taken by Constantine (or in his name) were to have great significance for centuries to come. One of the main issues was the shape that Christian churches were to take, since there was not, apparently, a tradition of monumental church buildings when Constantine decided to help the Christian church build a series of truly spectacular structures. The main form that these churches took was that of the basilica, a multipurpose rectangular structure, based ultimately on the earlier Greek stoa, which could be found in most of the great cities of the empire. Christianity, unlike classical polytheism, needed a large interior space for the celebration of its religious services, and the basilica aptly filled that need. We naturally do not know the degree to which the emperor was involved in the design of new churches, but it is tempting to connect this with the secular basilica that Constantine completed in the Roman forum (the so-called Basilica of Maxentius) and the one he probably built in Trier, in connection with his residence in the city at a time when he was still caesar.
|
69 |
'''
|
|
|
|
|
70 |
|
71 |
# 4️⃣ Generate, display, and save audio files in a loop.
|
72 |
generator = pipeline(
|
73 |
-
text, voice='af_bella',
|
74 |
speed=1, split_pattern=r'\n+'
|
75 |
)
|
76 |
for i, (gs, ps, audio) in enumerate(generator):
|
|
|
23 |
| Model | Published | Training Data | Compute (A100 80GB) | Released Voices | Released Langs |
|
24 |
| ----- | --------- | ------------- | ------------------- | --------------- | -------------- |
|
25 |
| v0.19 | 2024 Dec 25 | <100 hrs | 500 hrs @ $400 | 10 | 1 |
|
26 |
+
| **v1.0** | 2025 Jan 27 | Few hundred hrs | 1000 hrs @ $1000 | [31+](https://huggingface.co/hexgrad/Kokoro-82M/blob/main/VOICES.md) | [3+](https://huggingface.co/hexgrad/Kokoro-82M/blob/main/VOICES.md) |
|
27 |
|
28 |
Training is continuous. The v0.19 model was produced "on the way" to the v1.0 model, so the Compute footprints overlap.
|
29 |
|
|
|
46 |
The following can be run in a single cell on [Google Colab](https://colab.research.google.com/).
|
47 |
```py
|
48 |
# 1️⃣ Install kokoro
|
49 |
+
!pip install -q kokoro>=0.2.3 soundfile
|
50 |
# 2️⃣ Install espeak, used for out-of-dictionary fallback
|
51 |
!apt-get -qq -y install espeak-ng > /dev/null 2>&1
|
52 |
# You can skip espeak installation, but OOD words will be skipped unless you provide a fallback
|
|
|
57 |
import soundfile as sf
|
58 |
# 🇺🇸 'a' => American English
|
59 |
# 🇬🇧 'b' => British English
|
60 |
+
# 🇫🇷 'f' => French fr-fr
|
61 |
+
# 🇮🇳 'h' => Hindi hi
|
62 |
pipeline = KPipeline(lang_code='a') # make sure lang_code matches voice
|
63 |
|
64 |
# The following text is for demonstration purposes only, unseen during training
|
|
|
69 |
|
70 |
These were to have an enormous impact, not only because they were associated with Constantine, but also because, as in so many other areas, the decisions taken by Constantine (or in his name) were to have great significance for centuries to come. One of the main issues was the shape that Christian churches were to take, since there was not, apparently, a tradition of monumental church buildings when Constantine decided to help the Christian church build a series of truly spectacular structures. The main form that these churches took was that of the basilica, a multipurpose rectangular structure, based ultimately on the earlier Greek stoa, which could be found in most of the great cities of the empire. Christianity, unlike classical polytheism, needed a large interior space for the celebration of its religious services, and the basilica aptly filled that need. We naturally do not know the degree to which the emperor was involved in the design of new churches, but it is tempting to connect this with the secular basilica that Constantine completed in the Roman forum (the so-called Basilica of Maxentius) and the one he probably built in Trier, in connection with his residence in the city at a time when he was still caesar.
|
71 |
'''
|
72 |
+
# text = 'Le dromadaire resplendissant déambulait tranquillement dans les méandres en mastiquant de petites feuilles vernissées.'
|
73 |
+
# text = 'ट्रांसपोर्टरों की हड़ताल लगातार पांचवें दिन जारी, दिसंबर से इलेक्ट्रॉनिक टोल कलेक्शनल सिस्टम'
|
74 |
|
75 |
# 4️⃣ Generate, display, and save audio files in a loop.
|
76 |
generator = pipeline(
|
77 |
+
text, voice='af_bella', # <= change voice here
|
78 |
speed=1, split_pattern=r'\n+'
|
79 |
)
|
80 |
for i, (gs, ps, audio) in enumerate(generator):
|