Skip to content

SOTA Open Source TTS

Available in win10 (#225)

spicysamaa2442902acee
diff --git a/install_env.bat b/install_env.bat
index c37d3af..abbe9c6 100644
--- a/install_env.bat
+++ b/install_env.bat
@@ -1,17 +1,16 @@
@echo off
chcp 65001
-:: 设置是否使用镜像站的标志,true 表示使用,false 表示不使用
set USE_MIRROR=true
echo use_mirror = %USE_MIRROR%
-
-set no_proxy="127.0.0.1, 0.0.0.0, localhost"
setlocal enabledelayedexpansion
cd /D "%~dp0"
set PATH="%PATH%";%SystemRoot%\system32
+echo %PATH%
+
:: 安装Miniconda
:: 检查是否有特殊字符
echo "%CD%"| findstr /R /C:"[!#\$%&()\*+,;<=>?@\[\]\^`{|}~\u4E00-\u9FFF ] " >nul && (
@@ -20,6 +19,7 @@ echo "%CD%"| findstr /R /C:"[!#\$%&()\*+,;<=>?@\[\]\^`{|}~\u4E00-\u9FFF ] " >nul
goto end
)
)
+
:: 解决跨驱动器安装问题
set TMP=%CD%\fishenv
set TEMP=%CD%\fishenv
@@ -35,6 +35,7 @@ set API_FLAG_PATH=%~dp0API_FLAGS.txt
set MINICONDA_DOWNLOAD_URL=https://mirrors.tuna.tsinghua.edu.cn/anaconda/miniconda/Miniconda3-py310_23.3.1-0-Windows-x86_64.exe
set MINICONDA_CHECKSUM=307194e1f12bbeb52b083634e89cc67db4f7980bd542254b43d3309eaf7cb358
set conda_exists=F
+
:: 确定是否要安装conda
call "%CONDA_ROOT_PREFIX%\_conda.exe" --version >nul 2>&1
if "%ERRORLEVEL%" EQU "0" set conda_exists=T
@@ -134,7 +135,6 @@ for %%p in (%packages%) do (
)
-
if not "!install_packages!"=="" (
echo.
echo 正在安装以下包: !install_packages!
diff --git a/run_cmd.bat b/run_cmd.bat
index 204af9a..d481e8f 100644
--- a/run_cmd.bat
+++ b/run_cmd.bat
@@ -8,45 +8,44 @@ cd /D "%~dp0"
set PATH="%PATH%";%SystemRoot%\system32
-:: 检查是否有特殊字符
+
echo "%CD%"| findstr /R /C:"[!#\$%&()\*+,;<=>?@\[\]\^`{|}~\u4E00-\u9FFF ] " >nul && (
echo.
- echo 当前路径中存在特殊字符,请使fish-speech的路径不含特殊字符后再运行。 && (
+ echo There are special characters in the current path, please make the path of fish-speech free of special characters before running. && (
goto end
)
)
-:: 解决跨驱动器安装问题
+
set TMP=%CD%\fishenv
set TEMP=%CD%\fishenv
-:: 取消激活已经激活的环境
+
(call conda deactivate && call conda deactivate && call conda deactivate) 2>nul
-:: 安装路径配置
+
set CONDA_ROOT_PREFIX=%cd%\fishenv\conda
set INSTALL_ENV_DIR=%cd%\fishenv\env
-:: 环境隔离
+
set PYTHONNOUSERSITE=1
set PYTHONPATH=
set PYTHONHOME=
set "CUDA_PATH=%INSTALL_ENV_DIR%"
set "CUDA_HOME=%CUDA_PATH%"
-:: 激活环境
+
call "%CONDA_ROOT_PREFIX%\condabin\conda.bat" activate "%INSTALL_ENV_DIR%"
-:: 检查环境是否成功激活
+
if errorlevel 1 (
echo.
- echo 环境激活失败
+ echo Environment activation failed.
goto end
) else (
echo.
- echo 环境激活成功
+ echo Environment activation succeeded.
)
-:: 进入cmd
cmd /k "%*"
:end
diff --git a/start.bat b/start.bat
index 25e28d7..ceeec00 100644
--- a/start.bat
+++ b/start.bat
@@ -6,7 +6,6 @@ set PYTHONPATH=%~dp0
set PYTHON_CMD=%cd%\fishenv\env\python
set API_FLAG_PATH=%~dp0API_FLAGS.txt
-:: 设置Hugging Face镜像源
set no_proxy="localhost, 127.0.0.1, 0.0.0.0"
set HF_ENDPOINT=https://hf-mirror.com
%PYTHON_CMD% .\tools\download_models.py
@@ -15,11 +14,10 @@ setlocal enabledelayedexpansion
set "API_FLAGS="
set "flags="
-:: 检查API_FLAG文件是否存在
+
if exist "%API_FLAG_PATH%" (
for /f "usebackq tokens=*" %%a in ("%API_FLAG_PATH%") do (
set "line=%%a"
- :: 去除行尾的反斜杠和空白字符,并且跳过以#开头的行
if not "!line:~0,1!"=="#" (
set "line=!line: =<SPACE>!"
set "line=!line:\=!"
@@ -31,26 +29,23 @@ if exist "%API_FLAG_PATH%" (
)
)
-:: 去除API_FLAGS变量最后的空格
+
if not "!API_FLAGS!"=="" set "API_FLAGS=!API_FLAGS:~0,-1!"
-:: 初始化 flags 变量
set "flags="
-:: 检查是否包含 --api 参数
echo !API_FLAGS! | findstr /C:"--api" >nul 2>&1
if !errorlevel! equ 0 (
echo.
- echo 启动HTTP API推理
+ echo Start HTTP API...
set "mode=api"
goto process_flags
)
-:: 检查是否包含 --infer 参数
echo !API_FLAGS! | findstr /C:"--infer" >nul 2>&1
if !errorlevel! equ 0 (
echo.
- echo 启动WebUI推理
+ echo Start WebUI Inference...
set "mode=infer"
goto process_flags
)
@@ -63,12 +58,10 @@ for %%p in (!API_FLAGS!) do (
)
)
-:: 去除 flags 变量开头的空格
if not "!flags!"=="" set "flags=!flags:~1!"
echo Debug: flags = !flags!
-:: 根据 mode 变量启动相应的推理
if "!mode!"=="api" (
%PYTHON_CMD% -m tools.api !flags!
) else if "!mode!"=="infer" (
@@ -76,9 +69,10 @@ if "!mode!"=="api" (
)
echo.
-echo 接下来启动页面
+echo Next launch the page...
%PYTHON_CMD% fish_speech\webui\manage.py
+
:end
endlocal
pause
diff --git a/tools/post_api.py b/tools/post_api.py
new file mode 100644
index 0000000..f702a6a
--- /dev/null
+++ b/tools/post_api.py
@@ -0,0 +1,120 @@
+import argparse
+import base64
+import json
+
+import pyaudio
+import requests
+
+
+def wav_to_base64(file_path):
+ with open(file_path, "rb") as wav_file:
+ wav_content = wav_file.read()
+ base64_encoded = base64.b64encode(wav_content)
+ return base64_encoded.decode("utf-8")
+
+
+def play_audio(audio_content, format, channels, rate):
+ p = pyaudio.PyAudio()
+ stream = p.open(format=format, channels=channels, rate=rate, output=True)
+ stream.write(audio_content)
+ stream.stop_stream()
+ stream.close()
+ p.terminate()
+
+
+if __name__ == "__main__":
+ parser = argparse.ArgumentParser(
+ description="Send a WAV file and text to a server and receive synthesized audio."
+ )
+
+ parser.add_argument(
+ "--url", "-u", type=str, required=True, help="URL of the server"
+ )
+ parser.add_argument(
+ "--text", "-t", type=str, required=True, help="Text to be synthesized"
+ )
+ parser.add_argument(
+ "--reference_audio", "-ra", type=str, required=True, help="Path to the WAV file"
+ )
+ parser.add_argument(
+ "--reference_text",
+ "-rt",
+ type=str,
+ required=True,
+ help="Reference text for voice synthesis",
+ )
+ parser.add_argument(
+ "--max_new_tokens", type=int, default=0, help="Maximum new tokens to generate"
+ )
+ parser.add_argument(
+ "--chunk_length", type=int, default=30, help="Chunk length for synthesis"
+ )
+ parser.add_argument(
+ "--top_p", type=float, default=0.7, help="Top-p sampling for synthesis"
+ )
+ parser.add_argument(
+ "--repetition_penalty",
+ type=float,
+ default=1.5,
+ help="Repetition penalty for synthesis",
+ )
+ parser.add_argument(
+ "--temperature", type=float, default=0.7, help="Temperature for sampling"
+ )
+ parser.add_argument(
+ "--speaker", type=str, default=None, help="Speaker ID for voice synthesis"
+ )
+ parser.add_argument("--format", type=str, default="wav", help="Audio format")
+ parser.add_argument(
+ "--streaming", type=bool, default=False, help="Enable streaming response"
+ )
+ parser.add_argument(
+ "--channels", type=int, default=1, help="Number of audio channels"
+ )
+ parser.add_argument("--rate", type=int, default=44100, help="Sample rate for audio")
+
+ args = parser.parse_args()
+
+ base64_audio = wav_to_base64(args.reference_audio)
+
+ data = {
+ "text": args.text,
+ "reference_text": args.reference_text,
+ "reference_audio": base64_audio,
+ "max_new_tokens": args.max_new_tokens,
+ "chunk_length": args.chunk_length,
+ "top_p": args.top_p,
+ "repetition_penalty": args.repetition_penalty,
+ "temperature": args.temperature,
+ "speaker": args.speaker,
+ "format": args.format,
+ "streaming": args.streaming,
+ }
+
+ response = requests.post(args.url, json=data, stream=args.streaming)
+
+ audio_format = pyaudio.paInt16 # Assuming 16-bit PCM format
+
+ if response.status_code == 200:
+ if args.streaming:
+ p = pyaudio.PyAudio()
+ stream = p.open(
+ format=audio_format, channels=args.channels, rate=args.rate, output=True
+ )
+ for chunk in response.iter_content(chunk_size=1024):
+ if chunk:
+ stream.write(chunk)
+ stream.stop_stream()
+ stream.close()
+ p.terminate()
+ else:
+ audio_content = response.content
+
+ with open("generated_audio.wav", "wb") as audio_file:
+ audio_file.write(audio_content)
+
+ play_audio(audio_content, audio_format, args.channels, args.rate)
+ print("Audio has been saved to 'generated_audio.wav'.")
+ else:
+ print(f"Request failed with status code {response.status_code}")
+ print(response.json())