coqui-ai-TTS Quickstart Cheatsheet

terry 05/09/2025
《coqui-ai-TTS Quickstart Cheatsheet》

1. Create Conda Environment

conda create -n coqui-ai-TTS python=3.10
conda activate coqui-ai-TTS

2. Install

  • Use pretrained models only
pip install coqui-tts
  • For training/development
git clone https://github.com/idiap/coqui-ai-TTS
cd coqui-ai-TTS
pip install -e .

3. Models

  • List available models
tts --list_models
  • Change model storage location (temporary)
export XDG_DATA_HOME="/www/coqui/models"
export TTS_HOME="/www/coqui/models"
echo $XDG_DATA_HOME
  • Permanent setup (Linux/macOS)
nano ~/.bashrc   # or ~/.zshrc
# add:
export XDG_DATA_HOME="/www/coqui/models"
export TTS_HOME="/www/coqui/models"

source ~/.bashrc   # or . ~/.zshrc

4. Voice Cloning

Python API

import torch
from TTS.api import TTS

device = "cuda" if torch.cuda.is_available() else "cpu"
api = TTS("tts_models/multilingual/multi-dataset/xtts_v2").to(device)

# Clone voice
api.tts_to_file(
  text="Hello world",
  speaker_wav=["my/cloning/audio.wav"],
  speaker="MySpeaker1",
  language="en",
)

# Reuse cloned voice
api.tts_to_file(
  text="Hello again",
  speaker="MySpeaker1",
  language="en",
)

CLI

# Clone voice
tts --model_name "tts_models/multilingual/multi-dataset/xtts_v2" \
    --text "你好世界" \
    --language_idx "zh" \
    --speaker_wav "my/cloning/audio.wav" \
    --speaker_idx "MySpeaker1"

# Reuse cloned voice
tts --model_name "tts_models/multilingual/multi-dataset/xtts_v2" \
    --text "你好世界" \
    --language_idx "zh" \
    --speaker_idx "MySpeaker1"

⚠️ For Chinese:

pip install pypinyin

5. Voice Conversion

Python API

from TTS.api import TTS

tts = TTS("voice_conversion_models/multilingual/vctk/freevc24").to("cuda")
tts.voice_conversion_to_file(
  source_wav="my/source.wav",
  target_wav="my/target.wav",
  file_path="output.wav"
)

CLI

tts --model_name "voice_conversion_models/multilingual/multi-dataset/openvoice_v2" \
    --source_wav "source.wav" \
    --target_wav "target.wav" \
    --out_path "output.wav"

Related articles