SpeakStream

 ____                   _     ____  _
/ ___| _ __   ___  __ _| | __/ ___|| |_ _ __ ___  __ _ _ __ ___
\___ \| '_ \ / _ \/ _` | |/ /\___ \| __| '__/ _ \/ _` | '_ ` _ \
 ___) | |_) |  __/ (_| |   <  ___) | |_| | |  __/ (_| | | | | | |
|____/| .__/ \___|\__,_|_|\_\|____/ \__|_|  \___|\__,_|_| |_| |_|
      |_|

A streaming text-to-speech library built on OpenAI's API. Feed tokens as they arrive and hear them spoken back in real time.

Features

• 🎧 Sentence-aware streaming turns tokens into speech as soon as sentences are complete.
• 🦢 Optional audio ducking lowers other application volumes while speech plays and can be toggled at runtime.
• 🗣️ Change voices easily using OpenAI's voice models (Alloy, Echo, Fable, Onyx, Nova, Shimmer, and new voices Ash, Coral, Sage).
• ⏩ Adjust playback speed on the fly.
• 🔇 Mute/unmute or stop speech instantly.
• 🔊 Automatic output device switching via the default-device-sink crate.
• ✅ Tick and error sounds for progress and failures.

Setup

Add SpeakStream to your Cargo.toml:

[dependencies]
speakstream = { path = "path/to/speakstream" }

Then build the library:

cargo build --release

Example

use speakstream::ss::SpeakStream;
use async_openai::types::Voice;

#[tokio::main]
async fn main() {
    let mut speak = SpeakStream::new(Voice::Ash, 1.0, true, false);
    speak.add_token("Hello, world!");
    speak.complete_sentence();
}

Runtime audio ducking

Audio ducking can be enabled when creating a new stream or toggled at runtime:

let mut speak = SpeakStream::new(Voice::Ash, 1.0, true, true);
assert!(speak.is_audio_ducking_enabled());
speak.set_audio_ducking_enabled(false);
// manually start ducking when the user begins talking
speak.start_audio_ducking();
// ...record microphone input...
speak.stop_audio_ducking();

License

MIT