diff --git a/README.md b/README.md
index 85c2a7bf..35e5ba11 100644
--- a/README.md
+++ b/README.md
@@ -19,12 +19,12 @@ Play ChatGPT and other LLM with Xiaomi AI Speaker
- [通义千问](https://help.aliyun.com/zh/dashscope/developer-reference/api-details)
## 获取小米音响DID
-系统和Shell|Linux *sh|Windows CMD用户|Windows PowerShell用户
--|-|-|-
-1、安装包|`pip install miservice_fork`|`pip install miservice_fork`|`pip install miservice_fork`
-2、设置变量|`export MI_USER=xxx`
`export MI_PASS=xxx`|`set MI_USER=xxx`
`set MI_PASS=xxx`|`$env:MI_USER="xxx"`
`$env:MI_PASS="xxx"`
-3、取得MI_DID|`micli list` |`micli list` |`micli list`
-4、设置MI_DID|`export MI_DID=xxx`| `set MI_DID=xxx`| `$env:MI_DID="xxx"`
+| 系统和Shell | Linux *sh | Windows CMD用户 | Windows PowerShell用户 |
+| ------------- | ---------------------------------------------- | -------------------------------------- | ---------------------------------------------- |
+| 1、安装包 | `pip install miservice_fork` | `pip install miservice_fork` | `pip install miservice_fork` |
+| 2、设置变量 | `export MI_USER=xxx`
`export MI_PASS=xxx` | `set MI_USER=xxx`
`set MI_PASS=xxx` | `$env:MI_USER="xxx"`
`$env:MI_PASS="xxx"` |
+| 3、取得MI_DID | `micli list` | `micli list` | `micli list` |
+| 4、设置MI_DID | `export MI_DID=xxx` | `set MI_DID=xxx` | `$env:MI_DID="xxx"` |
- 注意不同shell 对环境变量的处理是不同的，尤其是powershell赋值时，可能需要双引号来包括值。
- 如果获取did报错时，请更换一下无线网络，有很大概率解决问题。
@@ -146,42 +146,36 @@ ChatGLM [文档](http://open.bigmodel.cn/doc/api#chatglm_130b)
## 配置项说明
-| 参数 | 说明 | 默认值 | 可选值 |
-| ------------------------ | ------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------- |
-| hardware | 设备型号 | | |
-| account | 小爱账户 | | |
-| password | 小爱账户密码 | | |
-| openai_key | openai的apikey | | |
-| serpapi_api_key | serpapi的key 参考 [SerpAPI](https://serpapi.com/) | | |
-| glm_key | chatglm 的 apikey | | |
-| gemini_key | gemini 的 apikey [参考](https://makersuite.google.com/app/apikey) | | |
-| qwen_key | qwen 的 apikey [参考](https://help.aliyun.com/zh/dashscope/developer-reference/api-details) | | |
-| cookie | 小爱账户cookie （如果用上面密码登录可以不填） | | |
-| mi_did | 设备did | | |
-| use_command | 使用 MI command 与小爱交互 | `false` | |
-| mute_xiaoai | 快速停掉小爱自己的回答 | `true` | |
-| verbose | 是否打印详细日志 | `false` | |
-| bot | 使用的 bot 类型，目前支持 chatgptapi,newbing, qwen, gemini | `chatgptapi` | |
-| tts | 使用的 TTS 类型 | `mi` | `edge`、 `openai`、`azure`、`volc` |
-| tts_voice | TTS 的嗓音 | `zh-CN-XiaoxiaoNeural`(edge), `alloy`(openai), `zh-CN-XiaoxiaoMultilingualNeural`(azure) | |
-| prompt | 自定义prompt | `请用100字以内回答` | |
-| keyword | 自定义请求词列表 | `["请"]` | |
-| change_prompt_keyword | 更改提示词触发列表 | `["更改提示词"]` | |
-| start_conversation | 开始持续对话关键词 | `开始持续对话` | |
-| end_conversation | 结束持续对话关键词 | `结束持续对话` | |
-| stream | 使用流式响应，获得更快的响应 | `false` | |
-| proxy | 支持 HTTP 代理，传入 http proxy URL | "" | |
-| gpt_options | OpenAI API 的参数字典 | `{}` | |
-| bing_cookie_path | NewBing使用的cookie路径，参考[这里]获取 | 也可通过环境变量 `COOKIE_FILE` 设置 | |
-| bing_cookies | NewBing使用的cookie字典，参考[这里]获取 | | |
-| deployment_id | Azure OpenAI 服务的 deployment ID | 参考这个[如何找到deployment_id](https://github.com/yihong0618/xiaogpt/issues/347#issuecomment-1784410784) | |
-| api_base | 如果需要替换默认的api,或者使用Azure OpenAI 服务 | 例如：`https://abc-def.openai.azure.com/` | |
-| azure_tts_speech_key | Azure TTS key | null | |
-| azure_tts_service_region | Azure TTS 服务地区 | `eastasia` | [Regions - Speech service - Azure AI services](https://learn.microsoft.com/en-us/azure/ai-services/speech-service/regions) |
-| volc_accesskey | 火山引擎accesskey [参考](https://console.volcengine.com/iam/keymanage/) | | |
-| volc_secretkey | 火山引擎secretkey [参考](https://console.volcengine.com/iam/keymanage/) | | |
-| volc_tts_app | 火山引擎 TTS app 服务 [参考]( https://console.volcengine.com/sami/) | | |
-| volc_tts_speaker | 火山引擎 TTS speaker [参考]( https://www.volcengine.com/docs/6489/93478) | `zh_female_qingxin` | |
+| 参数 | 说明 | 默认值 | 可选值 |
+| --------------------- | ------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------- | ----------------------------------------------------- |
+| hardware | 设备型号 | | |
+| account | 小爱账户 | | |
+| password | 小爱账户密码 | | |
+| openai_key | openai的apikey | | |
+| serpapi_api_key | serpapi的key 参考 [SerpAPI](https://serpapi.com/) | | |
+| glm_key | chatglm 的 apikey | | |
+| gemini_key | gemini 的 apikey [参考](https://makersuite.google.com/app/apikey) | | |
+| qwen_key | qwen 的 apikey [参考](https://help.aliyun.com/zh/dashscope/developer-reference/api-details) | | |
+| cookie | 小爱账户cookie （如果用上面密码登录可以不填） | | |
+| mi_did | 设备did | | |
+| use_command | 使用 MI command 与小爱交互 | `false` | |
+| mute_xiaoai | 快速停掉小爱自己的回答 | `true` | |
+| verbose | 是否打印详细日志 | `false` | |
+| bot | 使用的 bot 类型，目前支持 chatgptapi,newbing, qwen, gemini | `chatgptapi` | |
+| tts | 使用的 TTS 类型 | `mi` | `edge`、 `openai`、`azure`、`volc`、`baidu`、`google` |
+| tts_options | TTS 参数字典，参考 [tetos](https://github.com/frostming/tetos) 获取可用参数 | | |
+| prompt | 自定义prompt | `请用100字以内回答` | |
+| keyword | 自定义请求词列表 | `["请"]` | |
+| change_prompt_keyword | 更改提示词触发列表 | `["更改提示词"]` | |
+| start_conversation | 开始持续对话关键词 | `开始持续对话` | |
+| end_conversation | 结束持续对话关键词 | `结束持续对话` | |
+| stream | 使用流式响应，获得更快的响应 | `false` | |
+| proxy | 支持 HTTP 代理，传入 http proxy URL | "" | |
+| gpt_options | OpenAI API 的参数字典 | `{}` | |
+| bing_cookie_path | NewBing使用的cookie路径，参考[这里]获取 | 也可通过环境变量 `COOKIE_FILE` 设置 | |
+| bing_cookies | NewBing使用的cookie字典，参考[这里]获取 | | |
+| deployment_id | Azure OpenAI 服务的 deployment ID | 参考这个[如何找到deployment_id](https://github.com/yihong0618/xiaogpt/issues/347#issuecomment-1784410784) | |
+| api_base | 如果需要替换默认的api,或者使用Azure OpenAI 服务 | 例如：`https://abc-def.openai.azure.com/` | |
[这里]: https://github.com/acheong08/EdgeGPT#getting-authentication-required
@@ -300,6 +294,7 @@ docker run -v :/config -p 9527:9527 -e XIAOGPT_HOSTNAME==1.0.2; python_version < \"3.11\"",
"idna>=2.8",
"sniffio>=1.1",
+ "typing-extensions>=4.1; python_version < \"3.11\"",
]
files = [
- {file = "anyio-3.7.1-py3-none-any.whl", hash = "sha256:91dee416e570e92c64041bd18b900d1d6fa78dff7048769ce5ac5ddad004fbb5"},
- {file = "anyio-3.7.1.tar.gz", hash = "sha256:44a3c9aba0f5defa43261a8b3efb97891f2bd7d804e0e1f56419befa1adfc780"},
+ {file = "anyio-4.3.0-py3-none-any.whl", hash = "sha256:048e05d0f6caeed70d731f3db756d35dcc1f35747c8c403364a8332c630441b8"},
+ {file = "anyio-4.3.0.tar.gz", hash = "sha256:f75253795a87df48568485fd18cdd2a3fa5c4f7c5be8e5e36637733fce06fed6"},
]
[[package]]
@@ -286,6 +287,20 @@ files = [
{file = "charset_normalizer-3.3.2-py3-none-any.whl", hash = "sha256:3e4d1f6587322d2788836a99c69062fbb091331ec940e02d12d179c1d53e25fc"},
]
+[[package]]
+name = "click"
+version = "8.1.7"
+requires_python = ">=3.7"
+summary = "Composable command line interface toolkit"
+groups = ["default"]
+dependencies = [
+ "colorama; platform_system == \"Windows\"",
+]
+files = [
+ {file = "click-8.1.7-py3-none-any.whl", hash = "sha256:ae74fb96c20a0277a1d615f1e4d73c8414f5a98db8b799a7931d1582f3390c28"},
+ {file = "click-8.1.7.tar.gz", hash = "sha256:ca9853ad459e787e2192211578cc907e7594e294c7ccc834310722b41b9ca6de"},
+]
+
[[package]]
name = "colorama"
version = "0.4.6"
@@ -327,17 +342,6 @@ files = [
{file = "dataclasses_json-0.6.3.tar.gz", hash = "sha256:35cb40aae824736fdf959801356641836365219cfe14caeb115c39136f775d2a"},
]
-[[package]]
-name = "decorator"
-version = "5.1.1"
-requires_python = ">=3.5"
-summary = "Decorators for Humans"
-groups = ["default"]
-files = [
- {file = "decorator-5.1.1-py3-none-any.whl", hash = "sha256:b8c3f85900b9dc423225913c5aace94729fe1fa9763b38939a95226f02d37186"},
- {file = "decorator-5.1.1.tar.gz", hash = "sha256:637996211036b6385ef91435e4fae22989472f9d571faba8927ba8253acbc330"},
-]
-
[[package]]
name = "distro"
version = "1.9.0"
@@ -466,19 +470,6 @@ files = [
{file = "frozenlist-1.4.1.tar.gz", hash = "sha256:c037a86e8513059a2613aaba4d817bb90b9d9b6b69aace3ce9c877e8c8ed402b"},
]
-[[package]]
-name = "google"
-version = "3.0.0"
-summary = "Python bindings to the Google search engine."
-groups = ["default"]
-dependencies = [
- "beautifulsoup4",
-]
-files = [
- {file = "google-3.0.0-py2.py3-none-any.whl", hash = "sha256:889cf695f84e4ae2c55fbc0cfdaf4c1e729417fa52ab1db0485202ba173e4935"},
- {file = "google-3.0.0.tar.gz", hash = "sha256:143530122ee5130509ad5e989f0512f7cb218b2d4eddbafbad40fd10e8d8ccbe"},
-]
-
[[package]]
name = "google-ai-generativelanguage"
version = "0.6.2"
@@ -580,6 +571,23 @@ files = [
{file = "google_auth_httplib2-0.2.0-py2.py3-none-any.whl", hash = "sha256:b65a0a2123300dd71281a7bf6e64d65a0759287df52729bdd1ae2e47dc311a3d"},
]
+[[package]]
+name = "google-cloud-texttospeech"
+version = "2.16.3"
+requires_python = ">=3.7"
+summary = "Google Cloud Texttospeech API client library"
+groups = ["default"]
+dependencies = [
+ "google-api-core[grpc]!=2.0.*,!=2.1.*,!=2.10.*,!=2.2.*,!=2.3.*,!=2.4.*,!=2.5.*,!=2.6.*,!=2.7.*,!=2.8.*,!=2.9.*,<3.0.0dev,>=1.34.1",
+ "google-auth!=2.24.0,!=2.25.0,<3.0.0dev,>=2.14.1",
+ "proto-plus<2.0.0dev,>=1.22.3",
+ "protobuf!=3.20.0,!=3.20.1,!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.19.5",
+]
+files = [
+ {file = "google-cloud-texttospeech-2.16.3.tar.gz", hash = "sha256:fabc315032d137da0710bb4c268734d336212d8fa8316b23b277dd3a84ce721c"},
+ {file = "google_cloud_texttospeech-2.16.3-py2.py3-none-any.whl", hash = "sha256:5d1e23f9270908a5d7ecf2af04105fbd3a7ddde60fe48506e397bd18c1ece499"},
+]
+
[[package]]
name = "google-generativeai"
version = "0.5.1"
@@ -1306,17 +1314,6 @@ files = [
{file = "protobuf-4.25.1.tar.gz", hash = "sha256:57d65074b4f5baa4ab5da1605c02be90ac20c8b40fb137d6a8df9f416b0d0ce2"},
]
-[[package]]
-name = "py"
-version = "1.11.0"
-requires_python = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*"
-summary = "library with cross-python path, ini-parsing, io, code, log facilities"
-groups = ["default"]
-files = [
- {file = "py-1.11.0-py2.py3-none-any.whl", hash = "sha256:607c53218732647dff4acdfcd50cb62615cedf612e72d1724fb1a0cc6405b378"},
- {file = "py-1.11.0.tar.gz", hash = "sha256:51c75c4126074b472f746a24399ad32f6053d1b34b68d2fa41e558e6f4a98719"},
-]
-
[[package]]
name = "pyasn1"
version = "0.5.1"
@@ -1342,21 +1339,6 @@ files = [
{file = "pyasn1_modules-0.3.0.tar.gz", hash = "sha256:5bd01446b736eb9d31512a30d46c1ac3395d676c6f3cafa4c03eb54b9925631c"},
]
-[[package]]
-name = "pycryptodome"
-version = "3.9.9"
-requires_python = ">=2.6, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*"
-summary = "Cryptographic library for Python"
-groups = ["default"]
-files = [
- {file = "pycryptodome-3.9.9-cp39-cp39-manylinux1_i686.whl", hash = "sha256:2a68df525b387201a43b27b879ce8c08948a430e883a756d6c9e3acdaa7d7bd8"},
- {file = "pycryptodome-3.9.9-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:a4599c0ca0fc027c780c1c45ed996d5bef03e571470b7b1c7171ec1e1a90914c"},
- {file = "pycryptodome-3.9.9-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:b4e6b269a8ddaede774e5c3adbef6bf452ee144e6db8a716d23694953348cd86"},
- {file = "pycryptodome-3.9.9-cp39-cp39-win32.whl", hash = "sha256:a199e9ca46fc6e999e5f47fce342af4b56c7de85fae893c69ab6aa17531fb1e1"},
- {file = "pycryptodome-3.9.9-cp39-cp39-win_amd64.whl", hash = "sha256:6e89bb3826e6f84501e8e3b205c22595d0c5492c2f271cbb9ee1c48eb1866645"},
- {file = "pycryptodome-3.9.9.tar.gz", hash = "sha256:910e202a557e1131b1c1b3f17a63914d57aac55cf9fb9b51644962841c3995c4"},
-]
-
[[package]]
name = "pydantic"
version = "2.5.3"
@@ -1500,16 +1482,6 @@ files = [
{file = "pyparsing-3.1.2.tar.gz", hash = "sha256:a1bac0ce561155ecc3ed78ca94d3c9378656ad4c94c1270de543f621420f94ad"},
]
-[[package]]
-name = "pytz"
-version = "2020.5"
-summary = "World timezone definitions, modern and historical"
-groups = ["default"]
-files = [
- {file = "pytz-2020.5-py2.py3-none-any.whl", hash = "sha256:16962c5fb8db4a8f63a26646d8886e9d769b6c511543557bc84e9569fb9a9cb4"},
- {file = "pytz-2020.5.tar.gz", hash = "sha256:180befebb1927b16f6b57101720075a984c019ac16b1b7575673bea42c6c3da5"},
-]
-
[[package]]
name = "pyyaml"
version = "6.0.1"
@@ -1639,20 +1611,6 @@ files = [
{file = "requests-2.31.0.tar.gz", hash = "sha256:942c5a758f98d790eaed1a29cb6eefc7ffb0d1cf7af05c3d2791656dbd6ad1e1"},
]
-[[package]]
-name = "retry"
-version = "0.9.2"
-summary = "Easy to use retry decorator."
-groups = ["default"]
-dependencies = [
- "decorator>=3.4.2",
- "py<2.0.0,>=1.4.26",
-]
-files = [
- {file = "retry-0.9.2-py2.py3-none-any.whl", hash = "sha256:ccddf89761fa2c726ab29391837d4327f819ea14d244c232a1d24c67a2f98606"},
- {file = "retry-0.9.2.tar.gz", hash = "sha256:f8bfa8b99b69c4506d6f5bd3b0aabf77f98cdb17f3c9fc3f5ca820033336fba4"},
-]
-
[[package]]
name = "rich"
version = "13.7.1"
@@ -1682,17 +1640,6 @@ files = [
{file = "rsa-4.9.tar.gz", hash = "sha256:e38464a49c6c85d7f1351b0126661487a7e0a14a50f1675ec50eb34d4f20ef21"},
]
-[[package]]
-name = "six"
-version = "1.16.0"
-requires_python = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*"
-summary = "Python 2 and 3 compatibility utilities"
-groups = ["default"]
-files = [
- {file = "six-1.16.0-py2.py3-none-any.whl", hash = "sha256:8abb2f1d86890a2dfb989f9a77cfcfd3e47c2a354b01111771326f8aa26e0254"},
- {file = "six-1.16.0.tar.gz", hash = "sha256:1e61c37477a1626458e36f7b1d82aa5c9b094fa4802892072e49de9c60c4c926"},
-]
-
[[package]]
name = "sniffio"
version = "1.3.0"
@@ -1784,6 +1731,26 @@ files = [
{file = "tenacity-8.2.3.tar.gz", hash = "sha256:5398ef0d78e63f40007c1fb4c0bff96e1911394d2fa8d194f77619c05ff6cc8a"},
]
+[[package]]
+name = "tetos"
+version = "0.1.0"
+requires_python = ">=3.8"
+summary = "Unified interface for multiple Text-to-Speech (TTS) providers"
+groups = ["default"]
+dependencies = [
+ "anyio>=4.3.0",
+ "azure-cognitiveservices-speech>=1.37.0",
+ "click>=8.1.7",
+ "edge-tts>=6.1.10",
+ "google-cloud-texttospeech>=2.16.3",
+ "mutagen>=1.47.0",
+ "openai>=1.20.0",
+]
+files = [
+ {file = "tetos-0.1.0-py3-none-any.whl", hash = "sha256:3a3d2a2a93c2d22f950fae10c179e88bdc1cd6ee5d812c7dd5bf14751653a667"},
+ {file = "tetos-0.1.0.tar.gz", hash = "sha256:fa537bb769ff4e54e3f808dd76c5d98373387a0a0c1119502e6998e00b4dcc75"},
+]
+
[[package]]
name = "tqdm"
version = "4.66.1"
@@ -1845,24 +1812,6 @@ files = [
{file = "urllib3-2.1.0.tar.gz", hash = "sha256:df7aa8afb0148fa78488e7899b2c59b5f4ffcfa82e6c54ccb9dd37c1d7b52d54"},
]
-[[package]]
-name = "volcengine"
-version = "1.0.136"
-summary = "The Volcengine SDK for Python"
-groups = ["default"]
-dependencies = [
- "google>=3.0.0",
- "protobuf>=3.18.3",
- "pycryptodome==3.9.9",
- "pytz==2020.5",
- "requests>=2.25.1",
- "retry==0.9.2",
- "six>=1.0",
-]
-files = [
- {file = "volcengine-1.0.136.tar.gz", hash = "sha256:7d88eb54d39b100855880bc830f73ee8769fc7341ae5909c8fe6a20ab438f32e"},
-]
-
[[package]]
name = "wcwidth"
version = "0.2.13"
diff --git a/pyproject.toml b/pyproject.toml
index fa04ea4e..32f9dcaf 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -18,7 +18,6 @@ dependencies = [
"rich",
"zhipuai>=2.0.1",
"httpx[socks]",
- "edge-tts>=6.1.3",
"EdgeGPT==0.1.26",
"langchain>=0.0.343",
"beautifulsoup4>=4.12.0",
@@ -26,9 +25,7 @@ dependencies = [
"google-generativeai",
"numexpr>=2.8.6",
"dashscope>=1.10.0",
- "azure-cognitiveservices-speech>=1.37.0",
- "multidict>=6.0.5",
- "volcengine>=1.0.136",
+ "tetos>=0.1.0",
]
license = {text = "MIT"}
dynamic = ["version", "optional-dependencies"]
diff --git a/requirements.txt b/requirements.txt
index ccc919ec..6a9289b9 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -4,7 +4,7 @@
aiohttp==3.9.5
aiosignal==1.3.1
annotated-types==0.6.0
-anyio==3.7.1
+anyio==4.3.0
async-timeout==4.0.3; python_version < "3.11"
attrs==23.2.0
azure-cognitiveservices-speech==1.37.0
@@ -13,21 +13,21 @@ bingimagecreator==0.5.0
cachetools==5.3.2
certifi==2024.2.2
charset-normalizer==3.3.2
+click==8.1.7
colorama==0.4.6; platform_system == "Windows"
dashscope==1.17.0
dataclasses-json==0.6.3
-decorator==5.1.1
distro==1.9.0
edge-tts==6.1.10
edgegpt==0.1.26
exceptiongroup==1.2.0; python_version < "3.11"
frozenlist==1.4.1
-google==3.0.0
google-ai-generativelanguage==0.6.2
google-api-core==2.15.0
google-api-python-client==2.125.0
google-auth==2.26.1
google-auth-httplib2==0.2.0
+google-cloud-texttospeech==2.16.3
google-generativeai==0.5.1
google-search-results==2.4.2
googleapis-common-protos==1.62.0
@@ -61,34 +61,29 @@ packaging==23.2
prompt-toolkit==3.0.43
proto-plus==1.23.0
protobuf==4.25.1
-py==1.11.0
pyasn1==0.5.1
pyasn1-modules==0.3.0
-pycryptodome==3.9.9
pydantic==2.5.3
pydantic-core==2.14.6
pygments==2.17.2
pyjwt==2.8.0
pyparsing==3.1.2; python_version > "3.0"
-pytz==2020.5
pyyaml==6.0.1
regex==2023.12.25
requests==2.31.0
-retry==0.9.2
rich==13.7.1
rsa==4.9
-six==1.16.0
sniffio==1.3.0
socksio==1.0.0
soupsieve==2.5
sqlalchemy==2.0.25
tenacity==8.2.3
+tetos==0.1.0
tqdm==4.66.1
typing-extensions==4.9.0
typing-inspect==0.9.0
uritemplate==4.1.1
urllib3==2.1.0
-volcengine==1.0.136
wcwidth==0.2.13
websockets==12.0
yarl==1.9.4
diff --git a/xiao_config.json.example b/xiao_config.json.example
index bbf5ac3c..97c8e057 100644
--- a/xiao_config.json.example
+++ b/xiao_config.json.example
@@ -14,7 +14,7 @@
"verbose": false,
"bot": "chatgptapi",
"tts": "mi",
- "edge_tts_voice": "zh-CN-XiaoxiaoNeural",
+ "tts_options": {},
"prompt": "请用100字以内回答",
"keyword": ["请"],
"change_prompt_keyword": ["更改提示词"],
@@ -26,11 +26,5 @@
"bing_cookie_path": "",
"bing_cookies": {},
"api_base": "https://abc-def.openai.azure.com/",
- "deployment_id": "",
- "azure_tts_speech_key": null,
- "azure_tts_service_region": "eastasia",
- "volc_accesskey": "",
- "volc_secretkey": "",
- "volc_tts_app": "",
- "volc_tts_speaker": "zh_male_chunhou",
-}
\ No newline at end of file
+ "deployment_id": ""
+}
diff --git a/xiaogpt/cli.py b/xiaogpt/cli.py
index ed5a87ed..e51ddef3 100644
--- a/xiaogpt/cli.py
+++ b/xiaogpt/cli.py
@@ -86,27 +86,9 @@ def main():
help="show info",
)
parser.add_argument(
- "--azure_tts_speech_key",
- dest="azure_tts_speech_key",
- help="if use azure tts",
- )
- parser.add_argument(
- "--azure_tts_service_region",
- dest="azure_tts_service_region",
- help="if use azure tts",
- )
- tts_group = parser.add_mutually_exclusive_group()
- tts_group.add_argument(
- "--enable_edge_tts",
- dest="tts",
- action="store_const",
- const="edge",
- help="if use edge tts",
- )
- tts_group.add_argument(
"--tts",
- help="tts type",
- choices=["mi", "edge", "openai", "azure"],
+ help="TTS provider",
+ choices=["mi", "edge", "openai", "azure", "google", "baidu", "volc"],
)
bot_group = parser.add_mutually_exclusive_group()
bot_group.add_argument(
@@ -190,9 +172,15 @@ def main():
options = parser.parse_args()
config = Config.from_options(options)
- miboy = MiGPT(config)
+ async def main(config: Config) -> None:
+ miboy = MiGPT(config)
+ try:
+ await miboy.run_forever()
+ finally:
+ await miboy.close()
+
loop = asyncio.get_event_loop()
- loop.run_until_complete(miboy.run_forever())
+ loop.run_until_complete(main(config))
if __name__ == "__main__":
diff --git a/xiaogpt/config.py b/xiaogpt/config.py
index 141d8360..4a7bf84f 100644
--- a/xiaogpt/config.py
+++ b/xiaogpt/config.py
@@ -33,15 +33,6 @@
# add more here
}
-EDGE_TTS_DICT = {
- "用英语": "en-US-AriaNeural",
- "用日语": "ja-JP-NanamiNeural",
- "用法语": "fr-BE-CharlineNeural",
- "用韩语": "ko-KR-SunHiNeural",
- "用德语": "de-AT-JonasNeural",
- # add more here
-}
-
DEFAULT_COMMAND = ("5-1", "5-5")
KEY_WORD = ("帮我", "请")
@@ -80,23 +71,11 @@ class Config:
start_conversation: str = "开始持续对话"
end_conversation: str = "结束持续对话"
stream: bool = False
- tts: Literal["mi", "edge", "azure", "openai"] = "mi"
- tts_voice: str | None = None
+ tts: Literal["mi", "edge", "azure", "openai", "baidu", "google", "volc"] = "mi"
+ tts_options: dict[str, Any] = field(default_factory=dict)
gpt_options: dict[str, Any] = field(default_factory=dict)
bing_cookie_path: str = ""
bing_cookies: dict | None = None
- azure_tts_speech_key: str | None = None
- azure_tts_service_region: str = "eastasia"
- volc_accesskey: str = os.getenv(
- "VOLC_ACCESSKEY", ""
- ) # https://console.volcengine.com/iam/keymanage/
- volc_secretkey: str = os.getenv("VOLC_SECRETKEY", "")
- volc_tts_app: str = os.getenv(
- "VOLC_TTS_APP", ""
- ) # https://console.volcengine.com/sami
- volc_tts_speaker: str = os.getenv(
- "VOLC_TTS_SPEAPER", "zh_female_qingxin"
- ) # https://www.volcengine.com/docs/6489/93478
def __post_init__(self) -> None:
if self.proxy:
@@ -121,8 +100,6 @@ def __post_init__(self) -> None:
raise Exception(
"Using GPT api needs openai API key, please google how to"
)
- if self.tts == "azure" and not self.azure_tts_speech_key:
- raise Exception("Using Azure TTS needs azure speech key")
@property
def tts_command(self) -> str:
diff --git a/xiaogpt/tts/__init__.py b/xiaogpt/tts/__init__.py
index fd5569b9..82bcc392 100644
--- a/xiaogpt/tts/__init__.py
+++ b/xiaogpt/tts/__init__.py
@@ -1,7 +1,5 @@
-from xiaogpt.tts.base import TTS as TTS
-from xiaogpt.tts.edge import EdgeTTS as EdgeTTS
-from xiaogpt.tts.mi import MiTTS as MiTTS
-from xiaogpt.tts.volc import VolcTTS as VolcTTS
-from xiaogpt.tts.azure import AzureTTS
+from xiaogpt.tts.base import TTS
+from xiaogpt.tts.mi import MiTTS
+from xiaogpt.tts.tetos import TetosTTS
-__all__ = ["TTS", "EdgeTTS", "MiTTS", "AzureTTS", "VolcTTS"]
+__all__ = ["TTS", "TetosTTS", "MiTTS"]
diff --git a/xiaogpt/tts/azure.py b/xiaogpt/tts/azure.py
deleted file mode 100644
index 8e356020..00000000
--- a/xiaogpt/tts/azure.py
+++ /dev/null
@@ -1,98 +0,0 @@
-from __future__ import annotations
-
-import logging
-import tempfile
-from pathlib import Path
-from typing import Optional
-
-import azure.cognitiveservices.speech as speechsdk
-
-from xiaogpt.tts.base import AudioFileTTS
-from xiaogpt.utils import calculate_tts_elapse
-
-logger = logging.getLogger(__name__)
-
-
-class AzureTTS(AudioFileTTS):
- voice_name = "zh-CN-XiaoxiaoMultilingualNeural"
-
- async def make_audio_file(self, query: str, text: str) -> tuple[Path, float]:
- output_file = tempfile.NamedTemporaryFile(
- suffix=".mp3", mode="wb", delete=False, dir=self.dirname.name
- )
-
- speech_synthesizer = self._build_speech_synthesizer(output_file.name)
- result: Optional[speechsdk.SpeechSynthesisResult] = (
- speech_synthesizer.speak_text_async(text).get()
- )
- if result is None:
- raise RuntimeError(
- f"Failed to get tts from azure with voice={self.voice_name}"
- )
- # Check result
- if result.reason == speechsdk.ResultReason.SynthesizingAudioCompleted:
- logger.debug("Speech synthesized for text [{}]".format(text))
-
- return Path(output_file.name), calculate_tts_elapse(text)
- elif result.reason == speechsdk.ResultReason.Canceled:
- cancellation_details = result.cancellation_details
- logger.warning(f"Speech synthesis canceled: {cancellation_details.reason}")
- if cancellation_details.reason == speechsdk.CancellationReason.Error:
- errmsg = f"Error details: {cancellation_details.error_details}"
- logger.error(errmsg)
- raise RuntimeError(errmsg)
- raise RuntimeError(f"Failed to get tts from azure with voice={self.voice_name}")
-
- def _build_speech_synthesizer(self, filename: str):
- speech_key = self.config.azure_tts_speech_key
- service_region = self.config.azure_tts_service_region
- if not speech_key:
- raise Exception("Azure tts need speech key")
- speech_config = speechsdk.SpeechConfig(
- subscription=speech_key, region=service_region
- )
- speech_config.set_speech_synthesis_output_format(
- speechsdk.SpeechSynthesisOutputFormat.Audio16Khz32KBitRateMonoMp3
- )
- if self.config.proxy:
- host, port, username, password = self._parse_proxy(self.config.proxy)
-
- if username and password:
- speech_config.set_proxy(
- hostname=host, port=port, username=username, password=password
- )
- else:
- speech_config.set_proxy(hostname=host, port=port)
-
- speech_config.speech_synthesis_voice_name = (
- self.config.tts_voice or self.voice_name
- )
- speech_synthesizer = speechsdk.SpeechSynthesizer(
- speech_config=speech_config,
- audio_config=speechsdk.audio.AudioOutputConfig(filename=filename), # type: ignore
- )
- return speech_synthesizer
-
- def _parse_proxy(self, proxy_str: str):
- proxy_str = proxy_str
- proxy_str_splited = proxy_str.split("://")
- proxy_type = proxy_str_splited[0]
- proxy_addr = proxy_str_splited[1]
-
- if proxy_type == "http":
- if "@" in proxy_addr:
- proxy_addr_splited = proxy_addr.split("@")
- proxy_auth = proxy_addr_splited[0]
- proxy_addr_netloc = proxy_addr_splited[1]
- proxy_auth_splited = proxy_auth.split(":")
- username = proxy_auth_splited[0]
- password = proxy_auth_splited[1]
- else:
- proxy_addr_netloc = proxy_addr
- username, password = None, None
-
- proxy_addr_netloc_splited = proxy_addr_netloc.split(":")
- host = proxy_addr_netloc_splited[0]
- port = int(proxy_addr_netloc_splited[1])
- return host, port, username, password
- raise NotImplementedError
diff --git a/xiaogpt/tts/base.py b/xiaogpt/tts/base.py
index 5c7800de..51ae09af 100644
--- a/xiaogpt/tts/base.py
+++ b/xiaogpt/tts/base.py
@@ -56,7 +56,7 @@ async def get_if_xiaoai_is_playing(self):
return is_playing
@abc.abstractmethod
- async def synthesize(self, query: str, text_stream: AsyncIterator[str]) -> None:
+ async def synthesize(self, lang: str, text_stream: AsyncIterator[str]) -> None:
"""Synthesize speech from a stream of text."""
raise NotImplementedError
@@ -87,20 +87,20 @@ def __init__(
self._start_http_server()
@abc.abstractmethod
- async def make_audio_file(self, query: str, text: str) -> tuple[Path, float]:
+ async def make_audio_file(self, lang: str, text: str) -> tuple[Path, float]:
"""Synthesize speech from text and save it to a file.
Return the file path and the duration of the audio in seconds.
The file path must be relative to the self.dirname.
"""
raise NotImplementedError
- async def synthesize(self, query: str, text_stream: AsyncIterator[str]) -> None:
+ async def synthesize(self, lang: str, text_stream: AsyncIterator[str]) -> None:
queue: asyncio.Queue[tuple[str, float]] = asyncio.Queue()
finished = asyncio.Event()
async def worker():
async for text in text_stream:
- path, duration = await self.make_audio_file(query, text)
+ path, duration = await self.make_audio_file(lang, text)
url = f"http://{self.hostname}:{self.port}/{path.name}"
await queue.put((url, duration))
finished.set()
diff --git a/xiaogpt/tts/edge.py b/xiaogpt/tts/edge.py
deleted file mode 100644
index 33fb6f63..00000000
--- a/xiaogpt/tts/edge.py
+++ /dev/null
@@ -1,32 +0,0 @@
-import tempfile
-from pathlib import Path
-
-import edge_tts
-
-from xiaogpt.config import EDGE_TTS_DICT
-from xiaogpt.tts.base import AudioFileTTS
-from xiaogpt.utils import find_key_by_partial_string
-
-
-class EdgeTTS(AudioFileTTS):
- default_voice = "zh-CN-XiaoxiaoNeural"
-
- async def make_audio_file(self, query: str, text: str) -> tuple[Path, float]:
- voice = (
- find_key_by_partial_string(EDGE_TTS_DICT, query)
- or self.config.tts_voice
- or self.default_voice
- )
- communicate = edge_tts.Communicate(text, voice, proxy=self.config.proxy)
- duration = 0
- with tempfile.NamedTemporaryFile(
- suffix=".mp3", mode="wb", delete=False, dir=self.dirname.name
- ) as f:
- async for chunk in communicate.stream():
- if chunk["type"] == "audio":
- f.write(chunk["data"])
- elif chunk["type"] == "WordBoundary":
- duration = (chunk["offset"] + chunk["duration"]) / 1e7
- if duration == 0:
- raise RuntimeError(f"Failed to get tts from edge with voice={voice}")
- return (Path(f.name), duration)
diff --git a/xiaogpt/tts/mi.py b/xiaogpt/tts/mi.py
index e4882437..164091b6 100644
--- a/xiaogpt/tts/mi.py
+++ b/xiaogpt/tts/mi.py
@@ -27,7 +27,7 @@ async def say(self, text: str) -> None:
f"{self.config.tts_command} {text}",
)
- async def synthesize(self, query: str, text_stream: AsyncIterator[str]) -> None:
+ async def synthesize(self, lang: str, text_stream: AsyncIterator[str]) -> None:
async for text in text_stream:
await self.say(text)
await self.wait_for_duration(calculate_tts_elapse(text))
diff --git a/xiaogpt/tts/openai.py b/xiaogpt/tts/openai.py
deleted file mode 100644
index 54c7ce90..00000000
--- a/xiaogpt/tts/openai.py
+++ /dev/null
@@ -1,46 +0,0 @@
-from __future__ import annotations
-
-import tempfile
-from pathlib import Path
-from typing import TYPE_CHECKING
-
-import httpx
-
-from xiaogpt.tts.base import AudioFileTTS
-from xiaogpt.utils import calculate_tts_elapse
-
-if TYPE_CHECKING:
- import openai
-
-
-class OpenAITTS(AudioFileTTS):
- default_voice = "alloy"
-
- async def make_audio_file(self, query: str, text: str) -> tuple[Path, float]:
- output_file = tempfile.NamedTemporaryFile(
- suffix=".mp3", mode="wb", delete=False, dir=self.dirname.name
- )
- httpx_kwargs = {}
- if self.config.proxy:
- httpx_kwargs["proxies"] = self.config.proxy
- async with httpx.AsyncClient(trust_env=True, **httpx_kwargs) as sess:
- client = self._make_openai_client(sess)
-
- resp = await client.audio.speech.create(
- model="tts-1",
- input=text,
- voice=self.config.tts_voice or self.default_voice,
- )
- resp.stream_to_file(output_file.name)
- return Path(output_file.name), calculate_tts_elapse(text)
-
- def _make_openai_client(self, sess: httpx.AsyncClient) -> openai.AsyncOpenAI:
- import openai
-
- api_base = self.config.api_base
- if api_base and api_base.rstrip("/").endswith("openai.azure.com"):
- raise NotImplementedError("TTS is not supported for Azure OpenAI")
- else:
- return openai.AsyncOpenAI(
- api_key=self.config.openai_key, http_client=sess, base_url=api_base
- )
diff --git a/xiaogpt/tts/tetos.py b/xiaogpt/tts/tetos.py
new file mode 100644
index 00000000..1178e050
--- /dev/null
+++ b/xiaogpt/tts/tetos.py
@@ -0,0 +1,56 @@
+from __future__ import annotations
+
+import tempfile
+from pathlib import Path
+
+from miservice import MiNAService
+from tetos.base import Speaker
+
+from xiaogpt.config import Config
+from xiaogpt.tts.base import AudioFileTTS
+
+
+class TetosTTS(AudioFileTTS):
+ def __init__(
+ self, mina_service: MiNAService, device_id: str, config: Config
+ ) -> None:
+ super().__init__(mina_service, device_id, config)
+ self.speaker = self._get_speaker()
+
+ def _get_speaker(self) -> Speaker:
+ from tetos.azure import AzureSpeaker
+ from tetos.baidu import BaiduSpeaker
+ from tetos.edge import EdgeSpeaker
+ from tetos.google import GoogleSpeaker
+ from tetos.openai import OpenAISpeaker
+ from tetos.volc import VolcSpeaker
+
+ options = self.config.tts_options
+ allowed_speakers: list[str] = []
+ for speaker in (
+ OpenAISpeaker,
+ EdgeSpeaker,
+ AzureSpeaker,
+ VolcSpeaker,
+ GoogleSpeaker,
+ BaiduSpeaker,
+ ):
+ if (name := speaker.__name__[:-7].lower()) == self.config.tts:
+ try:
+ return speaker(**options)
+ except TypeError as e:
+ raise ValueError(
+ f"{e}. Please add them via `tts_options` config"
+ ) from e
+ else:
+ allowed_speakers.append(name)
+ raise ValueError(
+ f"Unsupported TTS: {self.config.tts}, allowed: {','.join(allowed_speakers)}"
+ )
+
+ async def make_audio_file(self, lang: str, text: str) -> tuple[Path, float]:
+ output_file = tempfile.NamedTemporaryFile(
+ suffix=".mp3", mode="wb", delete=False, dir=self.dirname.name
+ )
+ duration = await self.speaker.synthesize(text, output_file.name, lang=lang)
+ return Path(output_file.name), duration
diff --git a/xiaogpt/tts/volc.py b/xiaogpt/tts/volc.py
deleted file mode 100644
index 78634ead..00000000
--- a/xiaogpt/tts/volc.py
+++ /dev/null
@@ -1,130 +0,0 @@
-from __future__ import annotations
-
-import logging
-import tempfile
-from pathlib import Path
-from typing import Optional
-import json
-import os
-import time
-import base64
-import threading
-import httpx
-
-from volcengine.ApiInfo import ApiInfo
-from volcengine.Credentials import Credentials
-from volcengine.ServiceInfo import ServiceInfo
-from volcengine.base.Service import Service
-
-
-from xiaogpt.tts.base import AudioFileTTS
-from xiaogpt.utils import calculate_tts_elapse
-
-logger = logging.getLogger(__name__)
-
-
-class VolcTTS(AudioFileTTS):
- def __init__(self, mina_service, device_id, config):
- super().__init__(mina_service, device_id, config)
- self.token = get_token(config)
- self.client = httpx.Client()
- logger.info("Initializing VolcTTS {self.token}")
-
- async def make_audio_file(self, query: str, text: str) -> tuple[Path, float]:
- tts_payload = json.dumps(
- {
- "text": text,
- "speaker": self.config.volc_tts_speaker,
- "audio_config": {
- "format": "mp3",
- "sample_rate": 24000,
- "speech_rate": 0,
- },
- }
- )
-
- req = {
- "appkey": self.config.volc_tts_app,
- "token": self.token,
- "namespace": "TTS",
- "payload": tts_payload,
- }
-
- resp = self.client.post("https://sami.bytedance.com/api/v1/invoke", json=req)
- try:
- sami_resp = resp.json()
- logger.info(f"volc sami_resp {resp.status_code}")
- if resp.status_code != 200:
- print(sami_resp)
- except:
- logger.error(f"Failed to get tts from volcengine with voice=zh {text}")
-
- if sami_resp["status_code"] == 20000000 and len(sami_resp["data"]) > 0:
- audio_data = base64.b64decode(sami_resp["data"])
- with tempfile.NamedTemporaryFile(
- suffix=".mp3", mode="wb", delete=False, dir=self.dirname.name
- ) as f:
- f.write(audio_data)
-
- return Path(f.name), calculate_tts_elapse(text)
-
-
-## fetch token and save it to file
-## it's aimed to reduce the request to volcengine
-## it'll throw error if token is requested too frequently (more than 1 times per minute)
-def get_token(config):
- token_file = Path.home() / ".volc.token"
- if not Path.exists(token_file):
- token = request_token_data(config)
- else:
- with open(token_file, "r") as f:
- token = json.load(f)
- if token["expires_at"] < time.time():
- token = request_token_data(config)
-
- if not Path.exists(token_file):
- with open(token_file, "w") as f:
- json.dump(token, f)
- return token["token"]
-
-
-def request_token_data(config):
- sami_service = SAMIService()
- sami_service.set_ak(config.volc_accesskey)
- sami_service.set_sk(config.volc_secretkey)
-
- req = {
- "appkey": config.volc_tts_app,
- "token_version": "volc-auth-v1",
- "expiration": 24 * 3600,
- }
- token = sami_service.common_json_handler("GetToken", req)
- logger.info(f"Got token from volcengine {token}")
- return token
-
-
-class SAMIService(Service):
- def __init__(self):
- self.service_info = ServiceInfo(
- "open.volcengineapi.com",
- {},
- Credentials("", "", "sami", "cn-north-1"),
- 10,
- 10,
- )
- self.api_info = {
- "GetToken": ApiInfo(
- "POST", "/", {"Action": "GetToken", "Version": "2021-07-27"}, {}, {}
- ),
- }
- super(SAMIService, self).__init__(self.service_info, self.api_info)
-
- def common_json_handler(self, api, body):
- params = dict()
- try:
- body = json.dumps(body)
- res = self.json(api, params, body)
- res_json = json.loads(res)
- return res_json
- except Exception as e:
- raise Exception(str(e))
diff --git a/xiaogpt/xiaogpt.py b/xiaogpt/xiaogpt.py
index b0e1e619..1b8b99c0 100644
--- a/xiaogpt/xiaogpt.py
+++ b/xiaogpt/xiaogpt.py
@@ -23,8 +23,7 @@
WAKEUP_KEYWORD,
Config,
)
-from xiaogpt.tts import TTS, EdgeTTS, MiTTS, AzureTTS, VolcTTS
-from xiaogpt.tts.openai import OpenAITTS
+from xiaogpt.tts import TTS, MiTTS, TetosTTS
from xiaogpt.utils import (
parse_cookie_string,
)
@@ -53,6 +52,9 @@ def __init__(self, config: Config):
self.log.debug(config)
self.mi_session = ClientSession()
+ async def close(self):
+ await self.mi_session.close()
+
async def poll_latest_ask(self):
async with ClientSession() as session:
session._cookie_jar = self.cookie_jar
@@ -78,16 +80,16 @@ async def poll_latest_ask(self):
# if you want force mute xiaoai, comment this line below.
await asyncio.sleep(1 - d)
- async def init_all_data(self, session):
- await self.login_miboy(session)
+ async def init_all_data(self):
+ await self.login_miboy()
await self._init_data_hardware()
self.mi_session.cookie_jar.update_cookies(self.get_cookie())
self.cookie_jar = self.mi_session.cookie_jar
self.tts # init tts
- async def login_miboy(self, session):
+ async def login_miboy(self):
account = MiAccount(
- session,
+ self.mi_session,
self.config.account,
self.config.password,
str(self.mi_token_home),
@@ -179,7 +181,7 @@ def need_ask_gpt(self, record):
return (
self.in_conversation
and not query.startswith(WAKEUP_KEYWORD)
- or query.startswith(tuple(self.config.keyword))
+ or query.lower().startswith(tuple(w.lower() for w in self.config.keyword))
)
def need_change_prompt(self, record):
@@ -225,7 +227,7 @@ async def get_latest_ask_from_xiaoai(self, session: ClientSession) -> dict | Non
return None
async def _retry(self):
- await self.init_all_data(self.mi_session)
+ await self.init_all_data()
def _get_last_query(self, data: dict) -> dict | None:
if d := data.get("data"):
@@ -258,16 +260,10 @@ async def do_tts(self, value):
@functools.cached_property
def tts(self) -> TTS:
- if self.config.tts == "edge":
- return EdgeTTS(self.mina_service, self.device_id, self.config)
- elif self.config.tts == "azure":
- return AzureTTS(self.mina_service, self.device_id, self.config)
- elif self.config.tts == "openai":
- return OpenAITTS(self.mina_service, self.device_id, self.config)
- elif self.config.tts == "volc":
- return VolcTTS(self.mina_service, self.device_id, self.config)
- else:
+ if self.config.tts == "mi":
return MiTTS(self.mina_service, self.device_id, self.config)
+ else:
+ return TetosTTS(self.mina_service, self.device_id, self.config)
async def wait_for_tts_finish(self):
while True:
@@ -347,7 +343,7 @@ async def wakeup_xiaoai(self):
)
async def run_forever(self):
- await self.init_all_data(self.mi_session)
+ await self.init_all_data()
task = asyncio.create_task(self.poll_latest_ask())
assert task is not None # to keep the reference to task, do not remove this
print(
@@ -389,6 +385,7 @@ async def run_forever(self):
print("问题：" + query + "？")
if not self.chatbot.has_history():
query = f"{query}，{self.config.prompt}"
+ query += "，并用本段话的language code作为开头，用|分隔，如：en-US|你好……"
if self.config.mute_xiaoai:
await self.stop_if_xiaoai_is_playing()
else:
@@ -404,7 +401,7 @@ async def run_forever(self):
print("小爱没回")
print(f"以下是 {self.chatbot.name} 的回答: ", end="")
try:
- await self.tts.synthesize(query, self.ask_gpt(query))
+ await self.speak(self.ask_gpt(query))
except Exception as e:
print(f"{self.chatbot.name} 回答出错 {str(e)}")
else:
@@ -412,3 +409,18 @@ async def run_forever(self):
if self.in_conversation:
print(f"继续对话, 或用`{self.config.end_conversation}`结束对话")
await self.wakeup_xiaoai()
+
+ async def speak(self, text_stream: AsyncIterator[str]) -> None:
+ text = await anext(text_stream)
+ # See if the first part contains language code(e.g. en-US|Hello world)
+ lang, _, first_chunk = text.rpartition("|")
+ if len(lang) > 7:
+ # It is not a legal language code, discard it
+ lang, first_chunk = "", text
+
+ async def gen(): # reconstruct the generator
+ yield first_chunk
+ async for text in text_stream:
+ yield text
+
+ await self.tts.synthesize(lang or "zh-CN", gen())