mkfischer/openedai-speech

OpenedAI Speech

An OpenAI API compatible text to speech server.

Compatible with the OpenAI audio/speech API
Serves the /v1/audio/speech endpoint
Not affiliated with OpenAI in any way, does not require an OpenAI API Key
A free, private, text-to-speech server with custom voice cloning

Full Compatibility:

tts-1: alloy, echo, fable, onyx, nova, and shimmer (configurable)
tts-1-hd: alloy, echo, fable, onyx, nova, and shimmer (configurable, uses OpenAI samples by default)
response_format: mp3, opus, aac, or flac
speed 0.25-4.0 (and more)

Details:

model 'tts-1' via piper tts (fast, can use cpu)
model 'tts-1-hd' via coqui-ai/TTS xtts_v2 voice cloning (fast, but requires around 4GB GPU VRAM)
Can be run without TTS/xtts_v2, entirely on cpu
Custom cloned voices can be used for tts-1-hd, just save a WAV file in the /voices/ directory
You can map your own piper voices and xtts_v2 speaker clones via the voice_to_speaker.yaml configuration file
Occasionally, certain words or symbols may sound incorrect, you can fix them with regex via pre_process_map.yaml

If you find a better voice match for tts-1 or tts-1-hd, please let me know so I can update the defaults.

Version: 0.8.0, 2024-03-23

Cleanup, docs update.

Version: 0.7.3, 2024-03-20

Allow different xtts versions per voice in voice_to_speaker.yaml, ex. xtts_v2.0.2
Quality: Fix xtts sample rate (24000 vs. 22050 for piper) and pops
use CUDA 12.2-base in Dockerfile

API Documentation

Installation instructions

You can run the server via docker like so (recommended):

docker compose up

If you want a minimal docker image with piper support only (900MB vs. 13.5GB, see: Dockerfile.min). You can edit the docker-compose.yml to easily change this.

Manual instructions:

# Install the Python requirements
pip install -r requirements.txt
# install ffmpeg and curl
sudo apt install ffmpeg curl
# Download the voice models:
# for tts-1
bash download_voices_tts-1.sh
# and for tts-1-hd
bash download_voices_tts-1-hd.sh

Usage

usage: speech.py [-h] [--piper_cuda] [--xtts_device XTTS_DEVICE] [--preload PRELOAD] [-P PORT]
                 [-H HOST]

OpenedAI Speech API Server

options:
  -h, --help            show this help message and exit
  --piper_cuda          Enable cuda for piper. Note: --cuda/onnxruntime-gpu is not working for me,
                        but cpu is fast enough (default: False)
  --xtts_device XTTS_DEVICE
                        Set the device for the xtts model. The special value of 'none' will use
                        piper for all models. (default: cuda)
  --preload PRELOAD     Preload a model (Ex. 'xtts' or 'xtts_v2.0.2'). By default it's loaded on
                        first use. (default: None)
  -P PORT, --port PORT  Server tcp port (default: 8000)
  -H HOST, --host HOST  Host to listen on, Ex. 0.0.0.0 (default: localhost)

Sample API Usage

You can use it like this:

curl http://localhost:8000/v1/audio/speech -H "Content-Type: application/json" -d '{
    "model": "tts-1",
    "input": "The quick brown fox jumped over the lazy dog.",
    "voice": "alloy",
    "response_format": "mp3",
    "speed": 1.0
  }' > speech.mp3

Or just like this:

curl http://localhost:8000/v1/audio/speech -H "Content-Type: application/json" -d '{
    "input": "The quick brown fox jumped over the lazy dog."}' > speech.mp3

Or like this example from the OpenAI Text to speech guide:

import openai

client = openai.OpenAI(
  # This part is not needed if you set these environment variables before import openai
  # export OPENAI_API_KEY=sk-11111111111
  # export OPENAI_BASE_URL=http://localhost:8000/v1
  api_key = "sk-111111111",
  base_url = "http://localhost:8000/v1",
)

with client.audio.speech.with_streaming_response.create(
  model="tts-1",
  voice="alloy",
  input="Today is a wonderful day to build something people love!"
) as response:
  response.stream_to_file("speech.mp3")

Custom Voices Howto

Custom voices should be mono 22050 hz sample rate WAV files with low noise (no background music, etc.) and not contain any partial words.Sample voices for xtts should be at least 6 seconds long, but they can be longer. However, longer samples do not always produce better results.

You can use FFmpeg to process your audio files and prepare them for xtts, here are some examples:

# convert a multi-channel audio file to mono, set sample rate to 22050 hz, trim to 6 seconds, and output as WAV file.
ffmpeg -i input.mp3 -ac 1 -ar 22050 -t 6 -y me.wav
# use a simple noise filter to clean up audio, and select a start time start for sampling.
ffmpeg -i input.wav -af "highpass=f=200, lowpass=f=3000" -ac 1 -ar 22050 -ss 00:13:26.2 -t 6 -y me.wav
# A more complex noise reduction setup, including volume adjustment
ffmpeg -i input.mkv -af "highpass=f=200, lowpass=f=3000, volume=5, afftdn=nf=25" -ac 1 -ar 22050 -ss 00:13:26.2 -t 6 -y me.wav

Once your WAV file is prepared, save it in the /voices/ directory and update the voice_to_speaker.yaml file with the new file name.

For example:

...
tts-1-hd:
  me:
    model: xtts_v2.0.2 # you can specify different xtts versions
    speaker: voices/me.wav # this could be you

mkfischer / openedai-speech