requirements.txt

# stream2sentence is the core of RealtimeTTS - it quickly converts streamed text into sentences for real-time synthesis
stream2sentence==0.3.1

# azure-cognitiveservices-speech is for AzureEngine
azure-cognitiveservices-speech==1.43.0

# coqui_tts is for CoquiEngine
coqui_tts==0.26.0

# elevenlabs is for ElevenlabsEngine
elevenlabs==1.57.0

# gtts is for GTTSEngine
gtts==2.5.4

# openai is for OpenAIEngine
openai==1.75.0

# pyttsx3 is for SystemEngine
pyttsx3==2.98

# edge-tts is for EdgeEngine
edge-tts==7.0.1

# kokoro is for KokoroEngine
kokoro==0.9.4

# snac is for OrpheusEngine
snac==1.2.1

# fugashi is to support japanese language for KokoroEngine
fugashi==1.4.0
jaconv==0.4.0
mojimoji==0.0.13
pyopenjtalk==0.4.1

# pypinyin is to support chinese language for KokoroEngine
pypinyin==0.54.0st
ordered_set==4.1.0
jieba==0.42.1
cn2an==0.5.23

# pyaudio is for playing chunks over output device
pyaudio==0.2.14

# pydub is used to convert chunks from mp3 to pcm (for openai tts)
pydub==0.25.1

# resampy is used to resample from the tts to the target device sample rate 
resampy==0.4.3