# py3-TTS-Wrapper
[](https://badge.fury.io/py/py3-tts-wrapper)
[](https://codecov.io/gh/willwade/py3-tts-wrapper)
> **Contributions are welcome! Check our [contribution guide](./CONTRIBUTING.md).**
_TTS-Wrapper_ simplifies using text-to-speech APIs by providing a unified interface across multiple services, allowing easy integration and manipulation of TTS capabilities.
> ℹ️ Full documentation is available at [https://willwade.github.io/tts-wrapper/](https://willwade.github.io/tts-wrapper/)
## Requirements
- Python 3.10 or higher
- System dependencies (see below)
- API credentials for online services
## Supported Services
- AWS Polly
- Google TTS
- Microsoft Azure TTS
- IBM Watson
- ElevenLabs
- Wit.Ai
- eSpeak-NG
- Play.HT
- AVSynth (macOS only)
- SAPI (Windows only)
- Sherpa-Onnx (NB: Means you can run any ONNX model you want - eg Piper or MMS models)
### Experimental (Not fully featured or in a state of WIP)
- PicoTTS
- UWP (WinRT) Speech system (win 10+)
## Features
- **Text to Speech**: Convert text into spoken audio.
- **SSML Support**: Use Speech Synthesis Markup Language to enhance speech synthesis.
- **Voice and Language Selection**: Customize the voice and language for speech synthesis.
- **Streaming and Direct Play**: Stream audio or play it directly.
- **Pause, Resume, and Stop Controls**: Manage audio playback dynamically.
- **File Output**: Save spoken audio to files in various formats.
- **Unified Voice handling** Get Voices across all TTS engines with alike keys
- **Volume, Pitch, and Rate Controls** Control volume, pitch and rate with unified methods
## Feature Matrix
| Engine | Platform | Online/Offline | SSML | Word Boundaries | Streaming | Playback Control | Callbacks |
|------------|--------------------|--------------------|------|-----------------|-----------|------------------|-----------|
| Polly | Linux/MacOS/Windows| Online | Yes | Yes | Yes | Yes | Full |
| Google | Linux/MacOS/Windows| Online | Yes | Yes | Yes | Yes | Full |
| Microsoft | Linux/MacOS/Windows| Online | Yes | Yes | Yes | Yes | Full |
| Watson | Linux/MacOS/Windows| Online | Yes | Yes | Yes | Yes | Full |
| ElevenLabs | Linux/MacOS/Windows| Online | No* | Yes | Yes | Yes | Full |
| Play.HT | Linux/MacOS/Windows| Online | No* | No** | Yes | Yes | Basic |
| Wit.Ai | Linux/MacOS/Windows| Online | No* | No** | Yes | Yes | Basic |
| eSpeak | Linux/MacOS | Offline | Yes | No** | Yes | Yes | Basic |
| AVSynth | MacOS | Offline | No | No** | Yes | Yes | Basic |
| SAPI | Windows | Offline | Yes | Yes | Yes | Yes | Full |
| UWP | Windows | Offline | Yes | Yes | Yes | Yes | Full |
| Sherpa-ONNX| Linux/MacOS/Windows| Offline | No | No** | Yes | Yes | Basic |
**Notes**:
- **SSML**: Entries marked with No* indicate that while the engine doesn't support SSML natively, the wrapper will automatically strip SSML tags and process the plain text.
- **Word Boundaries**: Entries marked with No** use an estimation-based timing system that may not be accurate for precise synchronization needs.
- **Callbacks**:
- "Full" supports accurate word-level timing callbacks, onStart, and onEnd events
- "Basic" supports onStart and onEnd events, with estimated word timings
- **Playback Control**: All engines support pause, resume, and stop functionality through the wrapper's unified interface
- All engines support the following core features:
- Voice selection (`set_voice`)
- Property control (rate, volume, pitch)
- File output (WAV, with automatic conversion to MP3/other formats)
- Streaming playback
- Audio device selection
### Core Methods Available
| Method | Description | Availability |
|--------------------------|----------------------------------------------|--------------|
| `speak()` | Direct speech playback | All engines |
| `speak_streamed()` | Streamed speech playback | All engines |
| `synth_to_file()` | Save speech to file | All engines |
| `pause()`, `resume()` | Playback control | All engines |
| `stop()` | Stop playback | All engines |
| `set_property()` | Control rate/volume/pitch | All engines |
| `get_voices()` | List available voices | All engines |
| `set_voice()` | Select voice | All engines |
| `connect()` | Register event callbacks | All engines |
| `check_credentials()` | Verify API credentials | Online engines|
| `set_output_device()` | Select audio output device | All engines |
---
## Installation
### Package Name Note
This package is published on PyPI as `py3-tts-wrapper` but installs as `tts-wrapper`. This is because it's a fork of the original `tts-wrapper` project with Python 3 support and additional features.
### System Dependencies
This project requires the following system dependencies on Linux:
```sh
sudo apt-get install portaudio19-dev
```
or MacOS, using [Homebrew](https://brew.sh)
```sh
brew install portaudio
```
For PicoTTS on Debian systems:
```sh
sudo apt-get install libttspico-utils
```
The `espeak` TTS functionality requires the `espeak-ng` C library to be installed on your system:
- **Ubuntu/Debian**: `sudo apt install espeak-ng`
- **macOS**: `brew install espeak-ng`
- **Windows**: Download the binaries from https://espeak.sourceforge.net/
### Using pip
Install from PyPI with selected engines:
```sh
pip install "py3-tts-wrapper[google,microsoft,sapi,sherpaonnx,googletrans]"
```
Install from GitHub:
```sh
pip install "py3-tts-wrapper[google,microsoft,sapi,sherpaonnx,googletrans]@git+https://github.com/willwade/tts-wrapper"
```
Note: On macOS/zsh, you may need to use quotes:
```sh
pip install "py3-tts-wrapper[google,watson,polly,elevenlabs,microsoft,sherpaonnx]"
```
## Usage Guide
### Basic Usage
```python
from tts_wrapper import PollyClient
pollyClient = PollyClient(credentials=('aws_key_id', 'aws_secret_access_key'))
from tts_wrapper import PollyTTS
tts = PollyTTS(pollyClient)
ssml_text = tts.ssml.add('Hello, <break time="500ms"/> world!')
tts.speak(ssml_text)
```
You can use SSML or plain text
```python
from tts_wrapper import PollyClient
pollyClient = PollyClient(credentials=('aws_key_id', 'aws_secret_access_key'))
from tts_wrapper import PollyTTS
tts = PollyTTS(pollyClient)
tts.speak('Hello world')
```
For a full demo see the examples folder. You'll need to fill out the credentials.json (or credentials-private.json). Use them from cd'ing into the examples folder.
Tips on gaining keys are below.
### Authorization
Each service uses different methods for authentication:
#### Polly
```python
from tts_wrapper import PollyTTS, PollyClient
client = PollyClient(credentials=('aws_region','aws_key_id', 'aws_secret_access_key'))
tts = PollyTTS(client)
```
#### Google
```python
from tts_wrapper import GoogleTTS, GoogleClient
client = GoogleClient(credentials=('path/to/creds.json'))
tts = GoogleTTS(client)
```
or pass the auth file as dict - so in memory
```python
from tts_wrapper import GoogleTTS, GoogleClient
with open(os.getenv("GOOGLE_SA_PATH"), "r") as file:
credentials_dict = json.load(file)
client = GoogleClient(credentials=os.getenv('GOOGLE_SA_PATH'))
client = GoogleClient(credentials=credentials_dict)]
```
#### Microsoft
```python
from tts_wrapper import MicrosoftTTS, MicrosoftClient
client = MicrosoftClient(credentials=('subscription_key','subscription_region'))
tts = MicrosoftTTS(client)
```
#### Watson
```python
from tts_wrapper import WatsonTTS, WatsonClient
client = WatsonClient(credentials=('api_key', 'region', 'instance_id'))
tts = WatsonTTS(client)
```
**Note** If you have issues with SSL certification try
```python
from tts_wrapper import WatsonTTS, WatsonClient
client = WatsonClient(credentials=('api_key', 'region', 'instance_id'),disableSSLVerification=True)
tts = WatsonTTS(client)
```
#### ElevenLabs
```python
from tts_wrapper import ElevenLabsTTS, ElevenLabsClient
client = ElevenLabsClient(credentials=('api_key'))
tts = ElevenLabsTTS(client)
```
- **Note**: ElevenLabs does not support SSML.
#### Wit.Ai
```python
from tts_wrapper import WitAiTTS, WitAiClient
client = WitAiClient(credentials=('token'))
tts = WitAiTTS(client)
```
#### Play.HT
```python
from tts_wrapper import PlayHTClient, PlayHTTTS
client = PlayHTClient(credentials=('api_key', 'user_id'))
tts = PlayHTTTS(client)
```
- **Note**: Play.HT does not support SSML, but we automatically strip SSML tags if present.
#### UWP
```python
from tts_wrapper import UWPTTS, UWPClient
client = UWPClient()
tts = UWPTTS(client)
```
#### eSpeak
```python
from tts_wrapper import eSpeakClient, eSpeakTTS
client = eSpeakClient()
tts = eSpeakTTS(client)
```
Note: Requires espeak-ng to be installed on your system.
#### SAPI (Windows)
```python
from tts_wrapper import SAPIClient, SAPITTS
client = SAPIClient()
tts = SAPITTS(client)
```
Note: Only available on Windows systems.
#### AVSynth (macOS)
```python
from tts_wrapper import AVSynthClient, AVSynthTTS
client = AVSynthClient()
tts = AVSynthTTS(client)
```
Note: Only available on macOS. Provides high-quality speech synthesis with word timing support and voice property control.
#### GoogleTrans
Uses the gTTS library.
```python
from tts_wrapper import GoogleTransClient, GoogleTransTTS
voice_id = "en-co.uk" # Example voice ID for UK English
client = GoogleTransClient(voice_id)
# Initialize the TTS engine
tts = GoogleTransTTS(client)
```
#### Sherpa-ONNX
You can provide blank model path and tokens path - and we will use a default location..
```python
from tts_wrapper import SherpaOnnxClient, SherpaOnnxTTS
client = SherpaOnnxClient(model_path=None, tokens_path=None)
tts = SherpaOnnxTTS(client)
```
Set a voice like
```python
# Find voices/langs availables
voices = tts.get_voices()
print("Available voices:", voices)
# Set the voice using ISO code
iso_code = "eng" # Example ISO code for the voice - also ID in voice details
tts.set_voice(iso_code)
```
and then use speak, speak_streamed etc..
You then can perform the following methods.
### Advanced Usage
#### SSML
Even if you don't use SSML features that much its wise to use the same syntax - so pass SSML not text to all engines
```python
ssml_text = tts.ssml.add('Hello world!')
```
#### Plain Text
If you want to keep things simple each engine will convert plain text to SSML if its not.
```python
tts.speak('Hello World!')
```
#### Speak
This will use the default audio output of your device to play the audio immediately
```python
tts.speak(ssml_text)
```
#### Check Credentials
This will check if the credentials are valid. Its only on the client object. Eg
```python
client = MicrosoftClient(
credentials=(os.getenv("MICROSOFT_TOKEN"), os.getenv("MICROSOFT_REGION"))
)
if client.check_credentials():
print("Credentials are valid.")
else:
print("Credentials are invalid."
```
NB: Each engine has a different way of checking credentials. If they dont have a supported the parent class will check get_voices. If you want to save calls just do a get_voices call.
#### Streaming and Playback Control
#### `pause_audio()`, `resume_audio()`, `stop_audio()`
These methods manage audio playback by pausing, resuming, or stopping it.
NB: Only to be used for speak_streamed
You need to make sure the optional dependency is included for this
```sh
pip install py3-tts-wrapper[controlaudio,google.. etc
```
then
```python
client = GoogleClient(..)
tts = GoogleTTS(client)
try:
text = "This is a pause and resume test. The text will be longer, depending on where the pause and resume works"
audio_bytes = tts.synth_to_bytes(text)
tts.load_audio(audio_bytes)
print("Play audio for 3 seconds")
tts.play(1)
tts.pause(8)
tts.resume()
time.sleep(6)
finally:
tts.cleanup()
```
- the pause and resume are in seconds from the start of the audio
- Please use the cleanup method to ensure the audio is stopped and the audio device is released
NB: to do this we use pyaudio. If you have issues with this you may need to install portaudio19-dev - particularly on linux
```sh
sudo apt-get install portaudio19-dev
```
#### File Output
```python
tts.synth_to_file(ssml_text, 'output.mp3', format='mp3')
```
there is also "synth" method which is legacy. Note we support saving as mp3, wav or flac.
```Python
tts.synth('<speak>Hello, world!</speak>', 'hello.mp3', format='mp3)
```
Note you can also stream - and save. Just note it saves at the end of streaming entirely..
```python
ssml_text = tts.ssml.add('Hello world!')
tts.speak_streamed(ssml_text,filepath,'wav')
```
#### Fetch Available Voices
```python
voices = tts.get_voices()
print(voices)
```
NB: All voices will have a id, dict of language_codes, name and gender. Just note not all voice engines provide gender
#### Voice Selection
```python
tts.set_voice(voice_id,lang_code=en-US)
```
e.g.
```python
tts.set_voice('en-US-JessaNeural','en-US')
```
Use the id - not a name
#### SSML
```python
ssml_text = tts.ssml.add('Hello, <break time="500ms"/> world!')
tts.speak(ssml_text)
```
#### Volume, Rate and Pitch Control
Set volume:
```python
tts.set_property("volume", "90")
text_read = f"The current volume is 90"
text_with_prosody = tts.construct_prosody_tag(text_read)
ssml_text = tts.ssml.add(text_with_prosody)
```
- Volume is set on a scale of 0 (silent) to 100 (maximum).
- The default volume is 100 if not explicitly specified.
Set rate:
```python
tts.set_property("rate", "slow")
text_read = f"The current rate is SLOW"
text_with_prosody = tts.construct_prosody_tag(text_read)
ssml_text = tts.ssml.add(text_with_prosody)
```
Speech Rate:
- Rate is controlled using predefined options:
- x-slow: Very slow speaking speed.
- slow: Slow speaking speed.
- medium (default): Normal speaking speed.
- fast: Fast speaking speed.
- x-fast: Very fast speaking speed.
- If not specified, the speaking rate defaults to medium.
Set pitch:
```python
tts.set_property("pitch", "high")
text_read = f"The current pitch is SLOW"
text_with_prosody = tts.construct_prosody_tag(text_read)
ssml_text = tts.ssml.add(text_with_prosody)
```
Pitch Control:
- Pitch is adjusted using predefined options that affect the vocal tone:
- x-low: Very deep pitch.
- low: Low pitch.
- medium (default): Normal pitch.
- high: High pitch.
- x-high: Very high pitch.
- If not explicitly set, the pitch defaults to medium.
Use the ```tts.ssml.clear_ssml()``` method to clear all entries from the ssml list
#### `set_property()`
This method allows setting properties like `rate`, `volume`, and `pitch`.
```python
tts.set_property("rate", "fast")
tts.set_property("volume", "80")
tts.set_property("pitch", "high")
```
#### `get_property()`
This method retrieves the value of properties such as `volume`, `rate`, or `pitch`.
```python
current_volume = tts.get_property("volume")
print(f"Current volume: {current_volume}")
```
#### Using callbacks on word-level boundaries
Note only **Polly, Microsoft, Google, ElevenLabs, UWP, SAPI and Watson** can do this **correctly**. We can't do this in anything else but we do do a estimated tonings for all other engines (ie elevenlabs, witAi and Piper)
```python
def my_callback(word: str, start_time: float, end_time: float):
duration = end_time - start_time
print(f"Word: {word}, Duration: {duration:.3f}s")
def on_start():
print('Speech started')
def on_end():
print('Speech ended')
try:
text = "Hello, This is a word timing test"
ssml_text = tts.ssml.add(text)
tts.connect('onStart', on_start)
tts.connect('onEnd', on_end)
tts.start_playback_with_callbacks(ssml_text, callback=my_callback)
except Exception as e:
print(f"Error: {e}")
```
and it will output
```bash
Speech started
Word: Hello, Duration: 0.612s
Word: , Duration: 0.212s
Word: This, Duration: 0.364s
Word: is, Duration: 0.310s
Word: a, Duration: 0.304s
Word: word, Duration: 0.412s
Word: timing, Duration: 0.396s
Word: test, Duration: 0.424s
Speech ended
```
#### `connect()`
This method allows registering callback functions for events like `onStart` or `onEnd`.
```python
def on_start():
print("Speech started")
tts.connect('onStart', on_start)
```
## Audio Output Methods
The wrapper provides several methods for audio output, each suited for different use cases:
### 1. Direct Playback
The simplest method - plays audio immediately:
```python
tts.speak("Hello world")
```
### 2. Streaming Playback
Recommended for longer texts - streams audio as it's being synthesized:
```python
tts.speak_streamed("This is a long text that will be streamed as it's synthesized")
```
### 3. File Output
Save synthesized speech to a file:
```python
tts.synth_to_file("Hello world", "output.wav")
```
### 4. Raw Audio Data
For advanced use cases where you need the raw audio data:
```python
# Get raw PCM audio data as bytes
audio_bytes = tts.synth_to_bytes("Hello world")
```
### Audio Format Notes
- All engines output WAV format by default
- For MP3 or other formats, use external conversion libraries like `pydub`:
```python
from pydub import AudioSegment
import io
# Get WAV data
audio_bytes = tts.synth_to_bytes("Hello world")
# Convert to MP3
wav_audio = AudioSegment.from_wav(io.BytesIO(audio_bytes))
wav_audio.export("output.mp3", format="mp3")
```
---
### Example Use Cases
#### 1. Saving Audio to a File
You can use the `synth_to_bytestream` method to synthesize audio in any supported format and save it directly to a file.
```python
# Synthesize text into a bytestream in MP3 format
bytestream = tts.synth_to_bytestream("Hello, this is a test", format="mp3")
# Save the audio bytestream to a file
with open("output.mp3", "wb") as f:
f.write(bytestream.read())
print("Audio saved to output.mp3")
```
**Explanation**:
- The method synthesizes the given text into audio in MP3 format.
- The `BytesIO` object is then written to a file using the `.read()` method of the `BytesIO` class.
#### 2. Real-Time Playback Using `sounddevice`
If you want to play the synthesized audio live without saving it to a file, you can use the `sounddevice` library to directly play the audio from the `BytesIO` bytestream.
```python
import sounddevice as sd
import numpy as np
# Synthesize text into a bytestream in WAV format
bytestream = tts.synth_to_bytestream("Hello, this is a live playback test", format="wav")
# Convert the bytestream back to raw PCM audio data for playback
audio_data = np.frombuffer(bytestream.read(), dtype=np.int16)
# Play the audio using sounddevice
sd.play(audio_data, samplerate=tts.audio_rate)
sd.wait()
print("Live playback completed")
```
**Explanation**:
- The method synthesizes the text into a `wav` bytestream.
- The bytestream is converted to raw PCM data using `np.frombuffer()`, which is then fed into the `sounddevice` library for live playback.
- `sd.play()` plays the audio in real-time, and `sd.wait()` ensures that the program waits until playback finishes.
### Manual Audio Control
For advanced use cases where you need direct control over audio playback, you can use the raw audio data methods:
```python
from tts_wrapper import AVSynthClient, AVSynthTTS
import numpy as np
import sounddevice as sd
# Initialize TTS
client = AVSynthClient()
tts = AVSynthTTS(client)
# Method 1: Direct playback of entire audio
def play_audio_stream(tts, text: str):
"""Play entire audio at once."""
# Get raw audio data
audio_data = tts.synth_to_bytes(text)
# Convert to numpy array for playback
samples = np.frombuffer(audio_data, dtype=np.int16)
# Play the audio
sd.play(samples, samplerate=tts.audio_rate)
sd.wait()
# Method 2: Chunked playback for more control
def play_audio_chunked(tts, text: str, chunk_size: int = 4096):
"""Process and play audio in chunks for more control."""
# Get raw audio data
audio_data = tts.synth_to_bytes(text)
# Create a continuous stream
stream = sd.OutputStream(
samplerate=tts.audio_rate,
channels=1, # Mono audio
dtype=np.int16
)
with stream:
# Process in chunks
for i in range(0, len(audio_data), chunk_size):
chunk = audio_data[i:i + chunk_size]
if len(chunk) % 2 != 0: # Ensure even size for 16-bit audio
chunk = chunk[:-1]
samples = np.frombuffer(chunk, dtype=np.int16)
stream.write(samples)
```
This manual control allows you to:
- Process audio data in chunks
- Implement custom audio processing
- Control playback timing
- Add effects or modifications to the audio
- Implement custom buffering strategies
The chunked playback method is particularly useful for:
- Real-time audio processing
- Custom pause/resume functionality
- Volume adjustment during playback
- Progress tracking
- Memory-efficient handling of long audio
**Note**: Manual audio control requires the `sounddevice` and `numpy` packages:
```sh
pip install sounddevice numpy
```
## Developer's Guide
### Setting up the Development Environment
#### Using Pipenv
1. Clone the repository:
```sh
git clone https://github.com/willwade/tts-wrapper.git
cd tts-wrapper
```
2. Install the package and system dependencies:
```sh
pip install .
```
To install optional dependencies, use:
```sh
pip install .[google, watson, polly, elevenlabs, microsoft]
```
This will install Python dependencies and system dependencies required for this project. Note that system dependencies will only be installed automatically on Linux.
#### Using UV
1. [Install UV](https://docs.astral.sh/uv/#getting-started)
```sh
pip install uv
```
2. Clone the repository:
```sh
git clone https://github.com/willwade/tts-wrapper.git
cd tts-wrapper
```
3. Install Python dependencies:
```sh
uv sync
```
4. Install system dependencies (Linux only):
```sh
uv run postinstall
```
**NOTE**: to get a requirements.txt file for the project use `uv export --format requirements-txt --all-extras --no-hashes` juat be warned that this will include all dependencies including dev ones.
## Release a new build
```sh
git tag -a v0.1.0 -m "Release 0.1.0"
git push origin v0.1.0
```
### Adding a New Engine to TTS Wrapper
This guide provides a step-by-step approach to adding a new engine to the existing Text-to-Speech (TTS) wrapper system.
#### Step 1: Create Engine Directory Structure
1. **Create a new folder** for your engine within the `engines` directory. Name this folder according to your engine, such as `witai` for Wit.ai.
Directory structure:
```
engines/witai/
```
2. **Create necessary files** within this new folder:
- `__init__.py` - Makes the directory a Python package.
- `client.py` - Handles all interactions with the TTS API.
- `engine.py` - Contains the TTS class that integrates with your abstract TTS system.
- `ssml.py` - Defines any SSML handling specific to this engine.
Final directory setup:
```
engines/
└── witai/
├── __init__.py
├── client.py
├── engine.py
└── ssml.py
```
#### Step 2: Implement Client Functionality in `client.py`
Implement authentication and necessary setup for API connection. This file should manage tasks such as sending synthesis requests and fetching available voices.
```python
class TTSClient:
def __init__(self, api_key):
self.api_key = api_key
# Setup other necessary API connection details here
def synth(self, text, options):
# Code to send a synthesis request to the TTS API
pass
def get_voices(self):
# Code to retrieve available voices from the TTS API
pass
```
#### Step 3: Define the TTS Engine in `engine.py`
This class should inherit from the abstract TTS class and implement required methods such as `get_voices` and `synth_to_bytes`.
```python
from .client import TTSClient
from your_tts_module.abstract_tts import AbstractTTS
class WitTTS(AbstractTTS):
def __init__(self, api_key):
super().__init__()
self.client = TTSClient(api_key)
def get_voices(self):
return self.client.get_voices()
def synth_to_bytes(self, text, format='wav'):
return self.client.synth(text, {'format': format})
```
#### Step 4: Implement SSML Handling in `ssml.py`
If the engine has specific SSML requirements or supports certain SSML tags differently, implement this logic here.
```python
from your_tts_module.abstract_ssml import BaseSSMLRoot, SSMLNode
class EngineSSML(BaseSSMLRoot):
def add_break(self, time='500ms'):
self.root.add(SSMLNode('break', attrs={'time': time}))
```
#### Step 5: Update `__init__.py`
Make sure the `__init__.py` file properly imports and exposes the TTS class and any other public classes or functions from your engine.
```python
from .engine import WitTTS
from .ssml import EngineSSML
```
#### NB: Credentials Files
You can store your credentials in either:
- `credentials.json` - For development
- `credentials-private.json` - For private credentials (should be git-ignored)
Example structure (do NOT commit actual credentials):
```json
{
"Polly": {
"region": "your-region",
"aws_key_id": "your-key-id",
"aws_access_key": "your-access-key"
},
"Microsoft": {
"token": "your-subscription-key",
"region": "your-region"
}
}
```
### Service-Specific Setup
#### AWS Polly
- [Create an AWS account](https://aws.amazon.com/free)
- [Set up IAM credentials](https://docs.aws.amazon.com/polly/latest/dg/setting-up.html)
- [Polly API Documentation](https://docs.aws.amazon.com/polly/latest/dg/API_Operations.html)
#### Microsoft Azure
- [Create an Azure account](https://azure.microsoft.com/free)
- [Create a Speech Service resource](https://docs.microsoft.com/azure/cognitive-services/speech-service/get-started)
- [Azure Speech Service Documentation](https://docs.microsoft.com/azure/cognitive-services/speech-service/rest-text-to-speech)
#### Google Cloud
- [Create a Google Cloud account](https://cloud.google.com/free)
- [Set up a service account](https://cloud.google.com/text-to-speech/docs/quickstart-client-libraries)
- [Google TTS Documentation](https://cloud.google.com/text-to-speech/docs)
#### IBM Watson
- [Create an IBM Cloud account](https://cloud.ibm.com/registration)
- [Create a Text to Speech service instance](https://cloud.ibm.com/catalog/services/text-to-speech)
- [Watson TTS Documentation](https://cloud.ibm.com/apidocs/text-to-speech)
#### ElevenLabs
- [Create an ElevenLabs account](https://elevenlabs.io/)
- [Get your API key](https://docs.elevenlabs.io/authentication)
- [ElevenLabs Documentation](https://docs.elevenlabs.io/)
#### Play.HT
- [Create a Play.HT account](https://play.ht/)
- [Get your API credentials](https://docs.play.ht/reference/api-getting-started)
- [Play.HT Documentation](https://docs.play.ht/)
#### Wit.AI
- [Create a Wit.ai account](https://wit.ai/)
- [Create a new app and get token](https://wit.ai/docs/quickstart)
- [Wit.ai Documentation](https://wit.ai/docs)
## License
This project is licensed under the [MIT License](./LICENSE).
Raw data
{
"_id": null,
"home_page": null,
"name": "py3-tts-wrapper",
"maintainer": null,
"docs_url": null,
"requires_python": "<3.13,>=3.10",
"maintainer_email": null,
"keywords": "elevenlabs, gTTS, mms, playht, polly, sapi, speech synthesis, text-to-speech, tts, witai",
"author": null,
"author_email": "Giulio Bottari <giuliobottari@gmail.com>, Will Wade <willwade@gmail.com>",
"download_url": "https://files.pythonhosted.org/packages/63/39/a309997dd802cfe08c521ccd5df51c69affe52cbc74dafd0eb8645353b8a/py3_tts_wrapper-0.9.24.tar.gz",
"platform": null,
"description": "# py3-TTS-Wrapper\n\n[](https://badge.fury.io/py/py3-tts-wrapper)\n[](https://codecov.io/gh/willwade/py3-tts-wrapper)\n\n> **Contributions are welcome! Check our [contribution guide](./CONTRIBUTING.md).**\n\n_TTS-Wrapper_ simplifies using text-to-speech APIs by providing a unified interface across multiple services, allowing easy integration and manipulation of TTS capabilities. \n\n > \u2139\ufe0f Full documentation is available at [https://willwade.github.io/tts-wrapper/](https://willwade.github.io/tts-wrapper/)\n\n## Requirements\n\n- Python 3.10 or higher\n- System dependencies (see below)\n- API credentials for online services\n\n## Supported Services\n\n- AWS Polly\n- Google TTS\n- Microsoft Azure TTS\n- IBM Watson\n- ElevenLabs\n- Wit.Ai \n- eSpeak-NG\n- Play.HT\n- AVSynth (macOS only)\n- SAPI (Windows only)\n- Sherpa-Onnx (NB: Means you can run any ONNX model you want - eg Piper or MMS models)\n\n### Experimental (Not fully featured or in a state of WIP)\n\n- PicoTTS\n- UWP (WinRT) Speech system (win 10+)\n\n## Features\n- **Text to Speech**: Convert text into spoken audio.\n- **SSML Support**: Use Speech Synthesis Markup Language to enhance speech synthesis.\n- **Voice and Language Selection**: Customize the voice and language for speech synthesis.\n- **Streaming and Direct Play**: Stream audio or play it directly.\n- **Pause, Resume, and Stop Controls**: Manage audio playback dynamically.\n- **File Output**: Save spoken audio to files in various formats.\n- **Unified Voice handling** Get Voices across all TTS engines with alike keys\n- **Volume, Pitch, and Rate Controls** Control volume, pitch and rate with unified methods\n\n\n## Feature Matrix\n\n| Engine | Platform | Online/Offline | SSML | Word Boundaries | Streaming | Playback Control | Callbacks |\n|------------|--------------------|--------------------|------|-----------------|-----------|------------------|-----------|\n| Polly | Linux/MacOS/Windows| Online | Yes | Yes | Yes | Yes | Full |\n| Google | Linux/MacOS/Windows| Online | Yes | Yes | Yes | Yes | Full |\n| Microsoft | Linux/MacOS/Windows| Online | Yes | Yes | Yes | Yes | Full |\n| Watson | Linux/MacOS/Windows| Online | Yes | Yes | Yes | Yes | Full |\n| ElevenLabs | Linux/MacOS/Windows| Online | No* | Yes | Yes | Yes | Full |\n| Play.HT | Linux/MacOS/Windows| Online | No* | No** | Yes | Yes | Basic |\n| Wit.Ai | Linux/MacOS/Windows| Online | No* | No** | Yes | Yes | Basic |\n| eSpeak | Linux/MacOS | Offline | Yes | No** | Yes | Yes | Basic |\n| AVSynth | MacOS | Offline | No | No** | Yes | Yes | Basic |\n| SAPI | Windows | Offline | Yes | Yes | Yes | Yes | Full |\n| UWP | Windows | Offline | Yes | Yes | Yes | Yes | Full |\n| Sherpa-ONNX| Linux/MacOS/Windows| Offline | No | No** | Yes | Yes | Basic |\n\n**Notes**:\n- **SSML**: Entries marked with No* indicate that while the engine doesn't support SSML natively, the wrapper will automatically strip SSML tags and process the plain text.\n- **Word Boundaries**: Entries marked with No** use an estimation-based timing system that may not be accurate for precise synchronization needs.\n- **Callbacks**: \n - \"Full\" supports accurate word-level timing callbacks, onStart, and onEnd events\n - \"Basic\" supports onStart and onEnd events, with estimated word timings\n- **Playback Control**: All engines support pause, resume, and stop functionality through the wrapper's unified interface\n- All engines support the following core features:\n - Voice selection (`set_voice`)\n - Property control (rate, volume, pitch)\n - File output (WAV, with automatic conversion to MP3/other formats)\n - Streaming playback\n - Audio device selection\n\n### Core Methods Available\n\n| Method | Description | Availability |\n|--------------------------|----------------------------------------------|--------------|\n| `speak()` | Direct speech playback | All engines |\n| `speak_streamed()` | Streamed speech playback | All engines |\n| `synth_to_file()` | Save speech to file | All engines |\n| `pause()`, `resume()` | Playback control | All engines |\n| `stop()` | Stop playback | All engines |\n| `set_property()` | Control rate/volume/pitch | All engines |\n| `get_voices()` | List available voices | All engines |\n| `set_voice()` | Select voice | All engines |\n| `connect()` | Register event callbacks | All engines |\n| `check_credentials()` | Verify API credentials | Online engines|\n| `set_output_device()` | Select audio output device | All engines |\n\n---\n\n## Installation\n\n### Package Name Note\n\nThis package is published on PyPI as `py3-tts-wrapper` but installs as `tts-wrapper`. This is because it's a fork of the original `tts-wrapper` project with Python 3 support and additional features.\n\n### System Dependencies\n\nThis project requires the following system dependencies on Linux:\n\n```sh\nsudo apt-get install portaudio19-dev\n```\n\nor MacOS, using [Homebrew](https://brew.sh)\n\n```sh\nbrew install portaudio\n```\n\nFor PicoTTS on Debian systems:\n\n```sh\nsudo apt-get install libttspico-utils\n```\n\nThe `espeak` TTS functionality requires the `espeak-ng` C library to be installed on your system:\n\n- **Ubuntu/Debian**: `sudo apt install espeak-ng`\n- **macOS**: `brew install espeak-ng`\n- **Windows**: Download the binaries from https://espeak.sourceforge.net/\n\n### Using pip\n\nInstall from PyPI with selected engines:\n```sh\npip install \"py3-tts-wrapper[google,microsoft,sapi,sherpaonnx,googletrans]\"\n```\n\nInstall from GitHub:\n```sh\npip install \"py3-tts-wrapper[google,microsoft,sapi,sherpaonnx,googletrans]@git+https://github.com/willwade/tts-wrapper\"\n```\n\nNote: On macOS/zsh, you may need to use quotes:\n```sh\npip install \"py3-tts-wrapper[google,watson,polly,elevenlabs,microsoft,sherpaonnx]\"\n```\n\n\n\n## Usage Guide\n\n### Basic Usage\n\n```python\nfrom tts_wrapper import PollyClient\npollyClient = PollyClient(credentials=('aws_key_id', 'aws_secret_access_key'))\n\nfrom tts_wrapper import PollyTTS\n\ntts = PollyTTS(pollyClient)\nssml_text = tts.ssml.add('Hello, <break time=\"500ms\"/> world!')\ntts.speak(ssml_text)\n```\n\nYou can use SSML or plain text\n\n```python\nfrom tts_wrapper import PollyClient\npollyClient = PollyClient(credentials=('aws_key_id', 'aws_secret_access_key'))\nfrom tts_wrapper import PollyTTS\n\ntts = PollyTTS(pollyClient)\ntts.speak('Hello world')\n```\n\nFor a full demo see the examples folder. You'll need to fill out the credentials.json (or credentials-private.json). Use them from cd'ing into the examples folder. \nTips on gaining keys are below.\n\n### Authorization\n\nEach service uses different methods for authentication:\n\n#### Polly\n\n```python\nfrom tts_wrapper import PollyTTS, PollyClient\nclient = PollyClient(credentials=('aws_region','aws_key_id', 'aws_secret_access_key'))\n\ntts = PollyTTS(client)\n```\n\n#### Google\n\n```python\nfrom tts_wrapper import GoogleTTS, GoogleClient\nclient = GoogleClient(credentials=('path/to/creds.json'))\n\ntts = GoogleTTS(client)\n```\nor pass the auth file as dict - so in memory\n\n```python\nfrom tts_wrapper import GoogleTTS, GoogleClient\n\nwith open(os.getenv(\"GOOGLE_SA_PATH\"), \"r\") as file:\n credentials_dict = json.load(file)\n\nclient = GoogleClient(credentials=os.getenv('GOOGLE_SA_PATH'))\nclient = GoogleClient(credentials=credentials_dict)]\n```\n\n#### Microsoft\n\n```python\nfrom tts_wrapper import MicrosoftTTS, MicrosoftClient\nclient = MicrosoftClient(credentials=('subscription_key','subscription_region'))\n\ntts = MicrosoftTTS(client)\n```\n\n#### Watson\n\n```python\nfrom tts_wrapper import WatsonTTS, WatsonClient\nclient = WatsonClient(credentials=('api_key', 'region', 'instance_id'))\n\ntts = WatsonTTS(client)\n```\n\n**Note** If you have issues with SSL certification try\n\n```python\nfrom tts_wrapper import WatsonTTS, WatsonClient\nclient = WatsonClient(credentials=('api_key', 'region', 'instance_id'),disableSSLVerification=True)\n\ntts = WatsonTTS(client)\n```\n\n#### ElevenLabs\n\n```python\nfrom tts_wrapper import ElevenLabsTTS, ElevenLabsClient\nclient = ElevenLabsClient(credentials=('api_key'))\ntts = ElevenLabsTTS(client)\n```\n\n- **Note**: ElevenLabs does not support SSML.\n\n#### Wit.Ai\n\n```python\nfrom tts_wrapper import WitAiTTS, WitAiClient\nclient = WitAiClient(credentials=('token'))\ntts = WitAiTTS(client)\n```\n\n#### Play.HT\n\n```python\nfrom tts_wrapper import PlayHTClient, PlayHTTTS\nclient = PlayHTClient(credentials=('api_key', 'user_id'))\ntts = PlayHTTTS(client)\n```\n\n- **Note**: Play.HT does not support SSML, but we automatically strip SSML tags if present.\n\n#### UWP\n\n```python\nfrom tts_wrapper import UWPTTS, UWPClient\nclient = UWPClient()\ntts = UWPTTS(client)\n```\n\n#### eSpeak\n\n```python\nfrom tts_wrapper import eSpeakClient, eSpeakTTS\n\nclient = eSpeakClient()\ntts = eSpeakTTS(client)\n```\n\nNote: Requires espeak-ng to be installed on your system.\n\n#### SAPI (Windows)\n\n```python\nfrom tts_wrapper import SAPIClient, SAPITTS\n\nclient = SAPIClient()\ntts = SAPITTS(client)\n```\n\nNote: Only available on Windows systems.\n\n#### AVSynth (macOS)\n\n```python\nfrom tts_wrapper import AVSynthClient, AVSynthTTS\n\nclient = AVSynthClient()\ntts = AVSynthTTS(client)\n```\n\nNote: Only available on macOS. Provides high-quality speech synthesis with word timing support and voice property control.\n\n#### GoogleTrans\n\nUses the gTTS library. \n\n```python\nfrom tts_wrapper import GoogleTransClient, GoogleTransTTS\nvoice_id = \"en-co.uk\" # Example voice ID for UK English\nclient = GoogleTransClient(voice_id)\n# Initialize the TTS engine\ntts = GoogleTransTTS(client)\n```\n\n#### Sherpa-ONNX\n\nYou can provide blank model path and tokens path - and we will use a default location.. \n\n```python\nfrom tts_wrapper import SherpaOnnxClient, SherpaOnnxTTS\nclient = SherpaOnnxClient(model_path=None, tokens_path=None)\ntts = SherpaOnnxTTS(client)\n```\n\nSet a voice like\n\n```python\n# Find voices/langs availables\nvoices = tts.get_voices()\nprint(\"Available voices:\", voices)\n\n# Set the voice using ISO code\niso_code = \"eng\" # Example ISO code for the voice - also ID in voice details\ntts.set_voice(iso_code)\n```\nand then use speak, speak_streamed etc.. \n\nYou then can perform the following methods.\n\n### Advanced Usage\n\n#### SSML\n\nEven if you don't use SSML features that much its wise to use the same syntax - so pass SSML not text to all engines\n\n```python\nssml_text = tts.ssml.add('Hello world!')\n```\n\n#### Plain Text\n\nIf you want to keep things simple each engine will convert plain text to SSML if its not.\n\n```python\ntts.speak('Hello World!')\n```\n\n#### Speak \n\nThis will use the default audio output of your device to play the audio immediately\n\n```python\ntts.speak(ssml_text)\n```\n\n#### Check Credentials\n\nThis will check if the credentials are valid. Its only on the client object. Eg\n\n```python\n\n client = MicrosoftClient(\n credentials=(os.getenv(\"MICROSOFT_TOKEN\"), os.getenv(\"MICROSOFT_REGION\"))\n )\n if client.check_credentials():\n print(\"Credentials are valid.\")\n else:\n print(\"Credentials are invalid.\"\n\n```\n\nNB: Each engine has a different way of checking credentials. If they dont have a supported the parent class will check get_voices. If you want to save calls just do a get_voices call.\n\n#### Streaming and Playback Control\n\n#### `pause_audio()`, `resume_audio()`, `stop_audio()`\nThese methods manage audio playback by pausing, resuming, or stopping it.\nNB: Only to be used for speak_streamed\n\nYou need to make sure the optional dependency is included for this\n\n```sh\npip install py3-tts-wrapper[controlaudio,google.. etc\n```\n\nthen\n\n```python\n\nclient = GoogleClient(..)\ntts = GoogleTTS(client)\ntry:\n text = \"This is a pause and resume test. The text will be longer, depending on where the pause and resume works\"\n audio_bytes = tts.synth_to_bytes(text)\n tts.load_audio(audio_bytes)\n print(\"Play audio for 3 seconds\")\n tts.play(1)\n tts.pause(8)\n tts.resume()\n time.sleep(6)\nfinally:\n tts.cleanup()\n\n```\n\n- the pause and resume are in seconds from the start of the audio\n- Please use the cleanup method to ensure the audio is stopped and the audio device is released\n\nNB: to do this we use pyaudio. If you have issues with this you may need to install portaudio19-dev - particularly on linux\n\n```sh\nsudo apt-get install portaudio19-dev\n```\n\n\n#### File Output\n\n```python\ntts.synth_to_file(ssml_text, 'output.mp3', format='mp3')\n```\nthere is also \"synth\" method which is legacy. Note we support saving as mp3, wav or flac. \n\n```Python\ntts.synth('<speak>Hello, world!</speak>', 'hello.mp3', format='mp3)\n```\nNote you can also stream - and save. Just note it saves at the end of streaming entirely..\n\n```python\nssml_text = tts.ssml.add('Hello world!')\n\ntts.speak_streamed(ssml_text,filepath,'wav')\n```\n\n\n#### Fetch Available Voices\n\n```python\nvoices = tts.get_voices()\nprint(voices)\n```\n\nNB: All voices will have a id, dict of language_codes, name and gender. Just note not all voice engines provide gender\n\n#### Voice Selection\n\n```python\ntts.set_voice(voice_id,lang_code=en-US)\n```\n\ne.g.\n\n```python\ntts.set_voice('en-US-JessaNeural','en-US')\n```\n\nUse the id - not a name\n\n#### SSML\n\n```python\nssml_text = tts.ssml.add('Hello, <break time=\"500ms\"/> world!')\ntts.speak(ssml_text)\n```\n\n#### Volume, Rate and Pitch Control\n\nSet volume:\n```python\ntts.set_property(\"volume\", \"90\")\ntext_read = f\"The current volume is 90\"\ntext_with_prosody = tts.construct_prosody_tag(text_read)\nssml_text = tts.ssml.add(text_with_prosody)\n```\n- Volume is set on a scale of 0 (silent) to 100 (maximum).\n- The default volume is 100 if not explicitly specified.\n\nSet rate:\n\n```python\ntts.set_property(\"rate\", \"slow\")\ntext_read = f\"The current rate is SLOW\"\ntext_with_prosody = tts.construct_prosody_tag(text_read)\nssml_text = tts.ssml.add(text_with_prosody)\n```\nSpeech Rate:\n- Rate is controlled using predefined options:\n - x-slow: Very slow speaking speed.\n - slow: Slow speaking speed.\n - medium (default): Normal speaking speed.\n - fast: Fast speaking speed.\n - x-fast: Very fast speaking speed.\n- If not specified, the speaking rate defaults to medium.\n\nSet pitch:\n```python\ntts.set_property(\"pitch\", \"high\")\ntext_read = f\"The current pitch is SLOW\"\ntext_with_prosody = tts.construct_prosody_tag(text_read)\nssml_text = tts.ssml.add(text_with_prosody)\n```\nPitch Control:\n- Pitch is adjusted using predefined options that affect the vocal tone:\n - x-low: Very deep pitch.\n - low: Low pitch.\n - medium (default): Normal pitch.\n - high: High pitch.\n - x-high: Very high pitch.\n- If not explicitly set, the pitch defaults to medium.\n\nUse the ```tts.ssml.clear_ssml()``` method to clear all entries from the ssml list\n\n#### `set_property()`\nThis method allows setting properties like `rate`, `volume`, and `pitch`.\n\n```python\ntts.set_property(\"rate\", \"fast\")\ntts.set_property(\"volume\", \"80\")\ntts.set_property(\"pitch\", \"high\")\n```\n\n#### `get_property()`\nThis method retrieves the value of properties such as `volume`, `rate`, or `pitch`.\n\n```python\ncurrent_volume = tts.get_property(\"volume\")\nprint(f\"Current volume: {current_volume}\")\n```\n\n\n#### Using callbacks on word-level boundaries\n\nNote only **Polly, Microsoft, Google, ElevenLabs, UWP, SAPI and Watson** can do this **correctly**. We can't do this in anything else but we do do a estimated tonings for all other engines (ie elevenlabs, witAi and Piper)\n\n```python\ndef my_callback(word: str, start_time: float, end_time: float):\n duration = end_time - start_time\n print(f\"Word: {word}, Duration: {duration:.3f}s\")\n\ndef on_start():\n print('Speech started')\n\ndef on_end():\n print('Speech ended')\n\ntry:\n text = \"Hello, This is a word timing test\"\n ssml_text = tts.ssml.add(text)\n tts.connect('onStart', on_start)\n tts.connect('onEnd', on_end)\n tts.start_playback_with_callbacks(ssml_text, callback=my_callback)\nexcept Exception as e:\n print(f\"Error: {e}\")\n```\n\nand it will output\n\n```bash\nSpeech started\nWord: Hello, Duration: 0.612s\nWord: , Duration: 0.212s\nWord: This, Duration: 0.364s\nWord: is, Duration: 0.310s\nWord: a, Duration: 0.304s\nWord: word, Duration: 0.412s\nWord: timing, Duration: 0.396s\nWord: test, Duration: 0.424s\nSpeech ended\n```\n\n#### `connect()`\nThis method allows registering callback functions for events like `onStart` or `onEnd`.\n\n```python\ndef on_start():\n print(\"Speech started\")\n\ntts.connect('onStart', on_start)\n```\n\n\n## Audio Output Methods\n\nThe wrapper provides several methods for audio output, each suited for different use cases:\n\n### 1. Direct Playback\n\nThe simplest method - plays audio immediately:\n```python\ntts.speak(\"Hello world\")\n```\n\n### 2. Streaming Playback\n\nRecommended for longer texts - streams audio as it's being synthesized:\n```python\ntts.speak_streamed(\"This is a long text that will be streamed as it's synthesized\")\n```\n\n### 3. File Output\n\nSave synthesized speech to a file:\n```python\ntts.synth_to_file(\"Hello world\", \"output.wav\")\n```\n\n### 4. Raw Audio Data\n\nFor advanced use cases where you need the raw audio data:\n```python\n# Get raw PCM audio data as bytes\naudio_bytes = tts.synth_to_bytes(\"Hello world\")\n```\n\n### Audio Format Notes\n\n- All engines output WAV format by default\n- For MP3 or other formats, use external conversion libraries like `pydub`:\n ```python\n from pydub import AudioSegment\n import io\n \n # Get WAV data\n audio_bytes = tts.synth_to_bytes(\"Hello world\")\n \n # Convert to MP3\n wav_audio = AudioSegment.from_wav(io.BytesIO(audio_bytes))\n wav_audio.export(\"output.mp3\", format=\"mp3\")\n ```\n\n---\n\n### Example Use Cases\n\n#### 1. Saving Audio to a File\n\nYou can use the `synth_to_bytestream` method to synthesize audio in any supported format and save it directly to a file.\n\n```python\n# Synthesize text into a bytestream in MP3 format\nbytestream = tts.synth_to_bytestream(\"Hello, this is a test\", format=\"mp3\")\n\n# Save the audio bytestream to a file\nwith open(\"output.mp3\", \"wb\") as f:\n f.write(bytestream.read())\n\nprint(\"Audio saved to output.mp3\")\n```\n\n**Explanation**:\n- The method synthesizes the given text into audio in MP3 format.\n- The `BytesIO` object is then written to a file using the `.read()` method of the `BytesIO` class.\n\n#### 2. Real-Time Playback Using `sounddevice`\n\nIf you want to play the synthesized audio live without saving it to a file, you can use the `sounddevice` library to directly play the audio from the `BytesIO` bytestream.\n\n```python\nimport sounddevice as sd\nimport numpy as np\n\n# Synthesize text into a bytestream in WAV format\nbytestream = tts.synth_to_bytestream(\"Hello, this is a live playback test\", format=\"wav\")\n\n# Convert the bytestream back to raw PCM audio data for playback\naudio_data = np.frombuffer(bytestream.read(), dtype=np.int16)\n\n# Play the audio using sounddevice\nsd.play(audio_data, samplerate=tts.audio_rate)\nsd.wait()\n\nprint(\"Live playback completed\")\n```\n\n**Explanation**:\n- The method synthesizes the text into a `wav` bytestream.\n- The bytestream is converted to raw PCM data using `np.frombuffer()`, which is then fed into the `sounddevice` library for live playback.\n- `sd.play()` plays the audio in real-time, and `sd.wait()` ensures that the program waits until playback finishes.\n\n### Manual Audio Control\n\nFor advanced use cases where you need direct control over audio playback, you can use the raw audio data methods:\n\n```python\nfrom tts_wrapper import AVSynthClient, AVSynthTTS\nimport numpy as np\nimport sounddevice as sd\n\n# Initialize TTS\nclient = AVSynthClient()\ntts = AVSynthTTS(client)\n\n# Method 1: Direct playback of entire audio\ndef play_audio_stream(tts, text: str):\n \"\"\"Play entire audio at once.\"\"\"\n # Get raw audio data\n audio_data = tts.synth_to_bytes(text)\n \n # Convert to numpy array for playback\n samples = np.frombuffer(audio_data, dtype=np.int16)\n \n # Play the audio\n sd.play(samples, samplerate=tts.audio_rate)\n sd.wait()\n\n# Method 2: Chunked playback for more control\ndef play_audio_chunked(tts, text: str, chunk_size: int = 4096):\n \"\"\"Process and play audio in chunks for more control.\"\"\"\n # Get raw audio data\n audio_data = tts.synth_to_bytes(text)\n \n # Create a continuous stream\n stream = sd.OutputStream(\n samplerate=tts.audio_rate,\n channels=1, # Mono audio\n dtype=np.int16\n )\n \n with stream:\n # Process in chunks\n for i in range(0, len(audio_data), chunk_size):\n chunk = audio_data[i:i + chunk_size]\n if len(chunk) % 2 != 0: # Ensure even size for 16-bit audio\n chunk = chunk[:-1]\n samples = np.frombuffer(chunk, dtype=np.int16)\n stream.write(samples)\n```\n\nThis manual control allows you to:\n- Process audio data in chunks\n- Implement custom audio processing\n- Control playback timing\n- Add effects or modifications to the audio\n- Implement custom buffering strategies\n\nThe chunked playback method is particularly useful for:\n- Real-time audio processing\n- Custom pause/resume functionality\n- Volume adjustment during playback\n- Progress tracking\n- Memory-efficient handling of long audio\n\n**Note**: Manual audio control requires the `sounddevice` and `numpy` packages:\n```sh\npip install sounddevice numpy\n```\n\n\n## Developer's Guide\n\n### Setting up the Development Environment\n\n#### Using Pipenv\n\n\n1. Clone the repository:\n ```sh\n git clone https://github.com/willwade/tts-wrapper.git\n cd tts-wrapper\n ```\n\n2. Install the package and system dependencies:\n ```sh\n pip install .\n ```\n\n To install optional dependencies, use:\n ```sh\n pip install .[google, watson, polly, elevenlabs, microsoft]\n ```\n\nThis will install Python dependencies and system dependencies required for this project. Note that system dependencies will only be installed automatically on Linux.\n\n#### Using UV\n\n1. [Install UV](https://docs.astral.sh/uv/#getting-started)\n ```sh\n pip install uv\n ```\n\n2. Clone the repository:\n ```sh\n git clone https://github.com/willwade/tts-wrapper.git\n cd tts-wrapper\n ```\n\n3. Install Python dependencies:\n ```sh\n uv sync\n ```\n\n4. Install system dependencies (Linux only):\n ```sh\n uv run postinstall\n ```\n\n\n**NOTE**: to get a requirements.txt file for the project use `uv export --format requirements-txt --all-extras --no-hashes` juat be warned that this will include all dependencies including dev ones.\n\n## Release a new build\n\n```sh\ngit tag -a v0.1.0 -m \"Release 0.1.0\"\ngit push origin v0.1.0\n```\n\n### Adding a New Engine to TTS Wrapper\n\nThis guide provides a step-by-step approach to adding a new engine to the existing Text-to-Speech (TTS) wrapper system.\n\n#### Step 1: Create Engine Directory Structure\n\n1. **Create a new folder** for your engine within the `engines` directory. Name this folder according to your engine, such as `witai` for Wit.ai.\n\n Directory structure:\n \n ```\n engines/witai/\n ```\n\n2. **Create necessary files** within this new folder:\n\n - `__init__.py` - Makes the directory a Python package.\n - `client.py` - Handles all interactions with the TTS API.\n - `engine.py` - Contains the TTS class that integrates with your abstract TTS system.\n - `ssml.py` - Defines any SSML handling specific to this engine.\n\n Final directory setup:\n\n ```\n engines/\n \u2514\u2500\u2500 witai/\n \u251c\u2500\u2500 __init__.py\n \u251c\u2500\u2500 client.py\n \u251c\u2500\u2500 engine.py\n \u2514\u2500\u2500 ssml.py\n ```\n\n#### Step 2: Implement Client Functionality in `client.py`\n\nImplement authentication and necessary setup for API connection. This file should manage tasks such as sending synthesis requests and fetching available voices.\n\n```python\nclass TTSClient:\n def __init__(self, api_key):\n self.api_key = api_key\n # Setup other necessary API connection details here\n\n def synth(self, text, options):\n # Code to send a synthesis request to the TTS API\n pass\n\n def get_voices(self):\n # Code to retrieve available voices from the TTS API\n pass\n```\n\n#### Step 3: Define the TTS Engine in `engine.py`\n\nThis class should inherit from the abstract TTS class and implement required methods such as `get_voices` and `synth_to_bytes`.\n\n```python\nfrom .client import TTSClient\nfrom your_tts_module.abstract_tts import AbstractTTS\n\nclass WitTTS(AbstractTTS):\n def __init__(self, api_key):\n super().__init__()\n self.client = TTSClient(api_key)\n\n def get_voices(self):\n return self.client.get_voices()\n\n def synth_to_bytes(self, text, format='wav'):\n return self.client.synth(text, {'format': format})\n```\n\n#### Step 4: Implement SSML Handling in `ssml.py`\n\nIf the engine has specific SSML requirements or supports certain SSML tags differently, implement this logic here.\n\n```python\nfrom your_tts_module.abstract_ssml import BaseSSMLRoot, SSMLNode\n\nclass EngineSSML(BaseSSMLRoot):\n def add_break(self, time='500ms'):\n self.root.add(SSMLNode('break', attrs={'time': time}))\n```\n\n#### Step 5: Update `__init__.py`\n\nMake sure the `__init__.py` file properly imports and exposes the TTS class and any other public classes or functions from your engine.\n\n```python\nfrom .engine import WitTTS\nfrom .ssml import EngineSSML\n```\n\n#### NB: Credentials Files\n\nYou can store your credentials in either:\n- `credentials.json` - For development\n- `credentials-private.json` - For private credentials (should be git-ignored)\n\nExample structure (do NOT commit actual credentials):\n```json\n{\n \"Polly\": {\n \"region\": \"your-region\",\n \"aws_key_id\": \"your-key-id\",\n \"aws_access_key\": \"your-access-key\"\n },\n \"Microsoft\": {\n \"token\": \"your-subscription-key\",\n \"region\": \"your-region\"\n }\n}\n```\n\n### Service-Specific Setup\n\n#### AWS Polly\n- [Create an AWS account](https://aws.amazon.com/free)\n- [Set up IAM credentials](https://docs.aws.amazon.com/polly/latest/dg/setting-up.html)\n- [Polly API Documentation](https://docs.aws.amazon.com/polly/latest/dg/API_Operations.html)\n\n#### Microsoft Azure\n- [Create an Azure account](https://azure.microsoft.com/free)\n- [Create a Speech Service resource](https://docs.microsoft.com/azure/cognitive-services/speech-service/get-started)\n- [Azure Speech Service Documentation](https://docs.microsoft.com/azure/cognitive-services/speech-service/rest-text-to-speech)\n\n#### Google Cloud\n- [Create a Google Cloud account](https://cloud.google.com/free)\n- [Set up a service account](https://cloud.google.com/text-to-speech/docs/quickstart-client-libraries)\n- [Google TTS Documentation](https://cloud.google.com/text-to-speech/docs)\n\n#### IBM Watson\n- [Create an IBM Cloud account](https://cloud.ibm.com/registration)\n- [Create a Text to Speech service instance](https://cloud.ibm.com/catalog/services/text-to-speech)\n- [Watson TTS Documentation](https://cloud.ibm.com/apidocs/text-to-speech)\n\n#### ElevenLabs\n- [Create an ElevenLabs account](https://elevenlabs.io/)\n- [Get your API key](https://docs.elevenlabs.io/authentication)\n- [ElevenLabs Documentation](https://docs.elevenlabs.io/)\n\n#### Play.HT\n- [Create a Play.HT account](https://play.ht/)\n- [Get your API credentials](https://docs.play.ht/reference/api-getting-started)\n- [Play.HT Documentation](https://docs.play.ht/)\n\n#### Wit.AI\n- [Create a Wit.ai account](https://wit.ai/)\n- [Create a new app and get token](https://wit.ai/docs/quickstart)\n- [Wit.ai Documentation](https://wit.ai/docs)\n\n## License\n\nThis project is licensed under the [MIT License](./LICENSE).\n\n",
"bugtrack_url": null,
"license": null,
"summary": "TTS-Wrapper makes it easier to use text-to-speech APIs by providing a unified and easy-to-use interface.",
"version": "0.9.24",
"project_urls": {
"Bug Tracker": "https://github.com/willwade/tts-wrapper/issues"
},
"split_keywords": [
"elevenlabs",
" gtts",
" mms",
" playht",
" polly",
" sapi",
" speech synthesis",
" text-to-speech",
" tts",
" witai"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "a76449bcc847943048698548bdd1f587c13002455e4817d7cae20e9b3e04865f",
"md5": "6c07dd60bec604343df584cce2851b76",
"sha256": "7c93689f8ca811846d55d85de7b9d6f4118e11fdc99010d35daafc9ddbffb7d6"
},
"downloads": -1,
"filename": "py3_tts_wrapper-0.9.24-py3-none-any.whl",
"has_sig": false,
"md5_digest": "6c07dd60bec604343df584cce2851b76",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<3.13,>=3.10",
"size": 848138,
"upload_time": "2025-02-20T07:22:55",
"upload_time_iso_8601": "2025-02-20T07:22:55.713706Z",
"url": "https://files.pythonhosted.org/packages/a7/64/49bcc847943048698548bdd1f587c13002455e4817d7cae20e9b3e04865f/py3_tts_wrapper-0.9.24-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "6339a309997dd802cfe08c521ccd5df51c69affe52cbc74dafd0eb8645353b8a",
"md5": "2279ba289de783832710064ebca223e5",
"sha256": "a7f42bf421c01c0e180a7bce0d5aaee6d2a1220fd72704301d54479e6e244b1f"
},
"downloads": -1,
"filename": "py3_tts_wrapper-0.9.24.tar.gz",
"has_sig": false,
"md5_digest": "2279ba289de783832710064ebca223e5",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<3.13,>=3.10",
"size": 6966292,
"upload_time": "2025-02-20T07:22:57",
"upload_time_iso_8601": "2025-02-20T07:22:57.777276Z",
"url": "https://files.pythonhosted.org/packages/63/39/a309997dd802cfe08c521ccd5df51c69affe52cbc74dafd0eb8645353b8a/py3_tts_wrapper-0.9.24.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-02-20 07:22:57",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "willwade",
"github_project": "tts-wrapper",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "py3-tts-wrapper"
}