返回提交历史
Modified
docs/media.md
+6
-2
Modified
g4f/Provider/__init__.py
+1
-1
Modified
g4f/Provider/audio/EdgeTTS.py
+3
-2
Modified
g4f/Provider/audio/__init__.py
+2
-1
Added
g4f/Provider/audio/gTTS.py
+77
-0
Modified
g4f/image/__init__.py
+1
-1
XFEstudio/gpt4free
feat: add gTTS provider and update EdgeTTS & media docs
- In **docs/media.md**: - Updated the import to include `gTTS` alongside `EdgeTTS`, `Gemini`, and `PollinationsAI`. - Changed the audio parameter for EdgeTTS from `"locale": "en-US"` to `"language": "en"`. - Added a new code example demonstrating how to use the gTTS provider and save the output as "google-tts.mp3". - In **g4f/Provider/__init__.py**: - Replaced the import of `EdgeTTS` with a wildcard import (`from .audio import *`) to include all audio providers. - In **g4f/Provider/audio/EdgeTTS.py**: - Added a new class attribute `model_id = "edge-tts"`. - Changed the voice selection logic to use `cls.model_id` instead of the hardcoded string "edge-tts". - Updated the filename generation to use `[cls.model_id]` instead of `[cls.default_model]`. - In **g4f/Provider/audio/__init__.py**: - Added an import for the new `gTTS` provider. - Added new file **g4f/Provider/audio/gTTS.py**: - Implements the gTTS provider using the `gtts` library. - Defines provider attributes (`label`, `working`, `model_id`, etc.) and generates audio using a similar structure to EdgeTTS. - In **g4f/image/__init__.py**: - Modified the `get_extension` function to extract the extension using `.lower().lstrip('.')` instead of slicing with `[1:]`.
099d7283
代码差异
6 个文件
+90
-7
@@ -33,7 +33,7 @@ asyncio.run(main())
33
33
```python
34
34
from g4f.client import Client
35
35
36
from g4f.Provider import EdgeTTS, Gemini, PollinationsAI
36
from g4f.Provider import gTTS, EdgeTTS, Gemini, PollinationsAI
37
37
38
38
client = Client(provider=PollinationsAI)
39
39
response = client.media.generate("Hello", audio={"voice": "alloy", "format": "mp3"})
@@ -48,8 +48,12 @@ response = client.media.generate("Hello", model="gemini-audio")
48
48
response.data[0].save("gemini.ogx")
49
49
50
50
client = Client(provider=EdgeTTS)
51
response = client.media.generate("Hello", audio={"locale": "en-US"})
51
response = client.media.generate("Hello", audio={"language": "en"})
52
52
response.data[0].save("edge-tts.mp3")
53
54
client = Client(provider=gTTS)
55
response = client.media.generate("Hello", audio={"language": "en"})
56
response.data[0].save("google-tts.mp3")
53
57
```
54
58
55
59
#### **Transcribe an Audio File:**
@@ -29,7 +29,7 @@ try:
29
29
except ImportError as e:
30
30
debug.error("MiniMax providers not loaded:", e)
31
31
try:
32
from .audio import EdgeTTS
32
from .audio import *
33
33
except ImportError as e:
34
34
debug.error("Audio providers not loaded:", e)
35
35
@@ -20,6 +20,7 @@ from ..helper import format_image_prompt
20
20
class EdgeTTS(AsyncGeneratorProvider, ProviderModelMixin):
21
21
label = "Edge TTS"
22
22
working = has_edge_tts
23
model_id = "edge-tts"
23
24
default_language = "en"
24
25
default_locale = "en-US"
25
26
default_format = "mp3"
@@ -45,7 +46,7 @@ class EdgeTTS(AsyncGeneratorProvider, ProviderModelMixin):
45
46
prompt = format_image_prompt(messages, prompt)
46
47
if not prompt:
47
48
raise ValueError("Prompt is empty.")
48
voice = audio.get("voice", model if model and model != "edge-tts" else None)
49
voice = audio.get("voice", model if model and model != cls.model_id else None)
49
50
if not voice:
50
51
voices = await VoicesManager.create()
51
52
if "locale" in audio:
@@ -62,7 +63,7 @@ class EdgeTTS(AsyncGeneratorProvider, ProviderModelMixin):
62
63
voice = random.choice(voices)["Name"]
63
64
64
65
format = audio.get("format", cls.default_format)
65
filename = get_filename([cls.default_model], prompt, f".{format}", prompt)
66
filename = get_filename([cls.model_id], prompt, f".{format}", prompt)
66
67
target_path = os.path.join(get_media_dir(), filename)
67
68
ensure_media_dir()
68
69
@@ -1 +1,2 @@
1
from .EdgeTTS import EdgeTTS
1
from .EdgeTTS import EdgeTTS
2
from .gTTS import gTTS
@@ -0,0 +1,77 @@
1
from __future__ import annotations
2
3
import os
4
import random
5
import asyncio
6
7
try:
8
from gtts import gTTS as gTTS_Service
9
has_gtts = True
10
except ImportError:
11
has_gtts = False
12
13
from ...typing import AsyncResult, Messages
14
from ...providers.response import AudioResponse
15
from ...image.copy_images import get_filename, get_media_dir, ensure_media_dir
16
from ..base_provider import AsyncGeneratorProvider, ProviderModelMixin
17
from ..helper import format_image_prompt
18
19
locals = {
20
"en-AU": ["English (Australia)", "en", "com.au"],
21
"en-GB": ["English (United Kingdom)", "en", "co.uk"],
22
"en-US": ["English (United States)", "en", "us"],
23
"en-CA": ["English (Canada)", "en", "ca"],
24
"en-IN": ["English (India)", "en", "co.in"],
25
"en-IE": ["English (Ireland)", "en", "ie"],
26
"en-ZA": ["English (South Africa)", "en", "co.za"],
27
"en-NG": ["English (Nigeria)", "en", "com.ng"],
28
"fr-CA": ["French (Canada)", "fr", "ca"],
29
"fr-FR": ["French (France)", "fr", "fr"],
30
"de-DE": ["German (Germany)", "de", "de"],
31
"zh-CN": ["Mandarin (China Mainland)", "zh-CN", "com"],
32
"zh-TW": ["Mandarin (Taiwan)", "zh-TW", "com"],
33
"pt-BR": ["Portuguese (Brazil)", "pt", "com.br"],
34
"pt-PT": ["Portuguese (Portugal)", "pt", "pt"],
35
"es-MX": ["Spanish (Mexico)", "es", "com.mx"],
36
"es-ES": ["Spanish (Spain)", "es", "es"],
37
"es-US": ["Spanish (United States)", "es", "us"],
38
}
39
models = {locale[0]: {"lang": locale[1], "tld": locale[2]} for locale in locals.values()}
40
41
class gTTS(AsyncGeneratorProvider, ProviderModelMixin):
42
label = "gTTS (Google Text-to-Speech)"
43
working = has_gtts
44
model_id = "google-tts"
45
default_language = "en"
46
default_tld = "com"
47
default_format = "mp3"
48
models = list(models.keys())
49
50
@classmethod
51
async def create_async_generator(
52
cls,
53
model: str,
54
messages: Messages,
55
prompt: str = None,
56
audio: dict = {},
57
**kwargs
58
) -> AsyncResult:
59
prompt = format_image_prompt(messages, prompt)
60
if not prompt:
61
raise ValueError("Prompt is empty.")
62
format = audio.get("format", cls.default_format)
63
filename = get_filename([cls.model_id], prompt, f".{format}", prompt)
64
target_path = os.path.join(get_media_dir(), filename)
65
ensure_media_dir()
66
67
gTTS_Service(
68
prompt,
69
**{
70
"lang": audio.get("language", cls.default_language),
71
"tld": audio.get("tld", cls.default_tld),
72
"slow": audio.get("slow", False),
73
**models.get(model, {})
74
}
75
).save(target_path)
76
77
yield AudioResponse(f"/media/{filename}", audio=audio, text=prompt)
@@ -79,7 +79,7 @@ def to_image(image: ImageType, is_svg: bool = False) -> Image:
79
79
80
80
def get_extension(filename: str) -> Optional[str]:
81
81
if '.' in filename:
82
ext = os.path.splitext(filename)[1][1:].lower()
82
ext = os.path.splitext(filename)[1].lower().lstrip('.')
83
83
return ext if ext in EXTENSIONS_MAP else None
84
84
return None
85
85