Merge 4f4ab53252 into 5c528f1d8e

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "tts"
-version = "0.25.5"
+version = "0.25.6"
 authors = ["Nolan Darilek <nolan@thewordnerd.info>"]
 repository = "https://github.com/ndarilek/tts-rs"
 description = "High-level Text-To-Speech (TTS) interface"
@@ -30,7 +30,14 @@ env_logger = "0.10"
 
 [target.'cfg(windows)'.dependencies]
 tolk = { version = "0.5", optional = true }
-windows = { version = "0.48", features = ["Foundation", "Foundation_Collections", "Media_Core", "Media_Playback", "Media_SpeechSynthesis", "Storage_Streams"] }
+windows = { version = "0.51", features = [
+    "Foundation",
+    "Foundation_Collections",
+    "Media_Core",
+    "Media_Playback",
+    "Media_SpeechSynthesis",
+    "Storage_Streams",
+] }
 
 [target.'cfg(target_os = "linux")'.dependencies]
 speech-dispatcher = { version = "0.16", default-features = false }
@@ -43,7 +50,16 @@ objc = { version = "0.2", features = ["exception"] }
 
 [target.wasm32-unknown-unknown.dependencies]
 wasm-bindgen = "0.2"
-web-sys = { version = "0.3", features = ["EventTarget", "SpeechSynthesis", "SpeechSynthesisErrorCode", "SpeechSynthesisErrorEvent", "SpeechSynthesisEvent", "SpeechSynthesisUtterance", "SpeechSynthesisVoice", "Window", ] }
+web-sys = { version = "0.3", features = [
+    "EventTarget",
+    "SpeechSynthesis",
+    "SpeechSynthesisErrorCode",
+    "SpeechSynthesisErrorEvent",
+    "SpeechSynthesisEvent",
+    "SpeechSynthesisUtterance",
+    "SpeechSynthesisVoice",
+    "Window",
+] }
 
 [target.'cfg(target_os="android")'.dependencies]
 jni = "0.21"
@@ -52,4 +68,4 @@ ndk-glue = "0.7"
 
 [package.metadata.docs.rs]
 no-default-features = true
-features = ["speech_dispatcher_0_11"]
+features = ["speech_dispatcher_0_11"]

src/backends/android.rs

@@ -250,6 +250,7 @@ impl Backend for Android {
             utterance_callbacks: true,
             voice: false,
             get_voice: false,
+            synthesize: false,
         }
     }
 
@@ -284,6 +285,10 @@ impl Backend for Android {
         }
     }
 
+    fn synthesize(&mut self, text: &str) -> Result<Vec<u8>, Error> {
+        unimplemented!();
+    }
+
     fn stop(&mut self) -> Result<(), Error> {
         let vm = Self::vm()?;
         let mut env = vm.get_env()?;

src/backends/appkit.rs

@@ -125,6 +125,10 @@ impl Backend for AppKit {
         Ok(None)
     }
 
+    fn synthesize(&mut self, text: &str) -> Result<Vec<u8>, Error> {
+        unimplemented!();
+    }
+
     fn stop(&mut self) -> Result<(), Error> {
         trace!("stop()");
         unsafe {

src/backends/av_foundation.rs

@@ -170,6 +170,7 @@ impl Backend for AvFoundation {
             voice: true,
             get_voice: false,
             utterance_callbacks: true,
+            synthesize: false,
         }
     }
 
@@ -206,6 +207,10 @@ impl Backend for AvFoundation {
         Ok(Some(UtteranceId::AvFoundation(utterance)))
     }
 
+    fn synthesize(&mut self, text: &str) -> Result<Vec<u8>, Error> {
+        unimplemented!();
+    }
+
     fn stop(&mut self) -> Result<(), Error> {
         trace!("stop()");
         unsafe {

src/backends/speech_dispatcher.rs

@@ -85,6 +85,7 @@ impl Backend for SpeechDispatcher {
             voice: true,
             get_voice: false,
             utterance_callbacks: true,
+            synthesize: false,
         }
     }
 
@@ -108,6 +109,10 @@ impl Backend for SpeechDispatcher {
         }
     }
 
+    fn synthesize(&mut self, text: &str) -> Result<Vec<u8>, Error> {
+        unimplemented!();
+    }
+
     fn stop(&mut self) -> Result<(), Error> {
         trace!("stop()");
         self.0.cancel()?;

src/backends/tolk.rs

@@ -39,6 +39,10 @@ impl Backend for Tolk {
         Ok(None)
     }
 
+    fn synthesize(&mut self, text: &str) -> Result<Vec<u8>, Error> {
+        unimplemented!();
+    }
+
     fn stop(&mut self) -> Result<(), Error> {
         trace!("stop()");
         self.0.silence();

src/backends/web.rs

@@ -59,6 +59,7 @@ impl Backend for Web {
             voice: true,
             get_voice: true,
             utterance_callbacks: true,
+            synthesize: false,
         }
     }
 
@@ -121,6 +122,10 @@ impl Backend for Web {
         }
     }
 
+    fn synthesize(&mut self, text: &str) -> Result<Vec<u8>, Error> {
+        unimplemented!();
+    }
+
     fn stop(&mut self) -> Result<(), Error> {
         trace!("stop()");
         if let Some(window) = web_sys::window() {

src/backends/winrt.rs

@@ -12,8 +12,11 @@ use windows::{
     Media::{
         Core::MediaSource,
         Playback::{MediaPlayer, MediaPlayerAudioCategory},
-        SpeechSynthesis::{SpeechSynthesizer, VoiceGender, VoiceInformation},
+        SpeechSynthesis::{
+            SpeechSynthesisStream, SpeechSynthesizer, VoiceGender, VoiceInformation,
+        },
     },
+    Storage::Streams::DataReader,
 };
 
 use crate::{Backend, BackendId, Error, Features, Gender, UtteranceId, Voice, CALLBACKS};
@@ -138,6 +141,20 @@ impl WinRt {
             voice: SpeechSynthesizer::DefaultVoice()?,
         })
     }
+
+    fn create_synthesis_stream(&mut self, text: &str) -> Result<SpeechSynthesisStream, Error> {
+        self.synth.Options()?.SetSpeakingRate(self.rate.into())?;
+        self.synth.Options()?.SetAudioPitch(self.pitch.into())?;
+        self.synth.Options()?.SetAudioVolume(self.volume.into())?;
+
+        self.synth.SetVoice(&self.voice)?;
+        let stream = self
+            .synth
+            .SynthesizeTextToStreamAsync(&text.into())?
+            .get()?;
+
+        Ok(stream)
+    }
 }
 
 impl Backend for WinRt {
@@ -155,6 +172,7 @@ impl Backend for WinRt {
             voice: true,
             get_voice: true,
             utterance_callbacks: true,
+            synthesize: true,
         }
     }
 
@@ -189,18 +207,12 @@ impl Backend for WinRt {
             }
         }
         if no_utterances {
-            self.synth.Options()?.SetSpeakingRate(self.rate.into())?;
-            self.synth.Options()?.SetAudioPitch(self.pitch.into())?;
-            self.synth.Options()?.SetAudioVolume(self.volume.into())?;
-            self.synth.SetVoice(&self.voice)?;
-            let stream = self
-                .synth
-                .SynthesizeTextToStreamAsync(&text.into())?
-                .get()?;
-            let content_type = stream.ContentType()?;
-            let source = MediaSource::CreateFromStream(&stream, &content_type)?;
-            self.player.SetSource(&source)?;
+            let stream = self.create_synthesis_stream(text)?;
+
+            let media_source = MediaSource::CreateFromStream(&stream, &stream.ContentType()?)?;
+            self.player.SetSource(&media_source)?;
             self.player.Play()?;
+
             let mut callbacks = CALLBACKS.lock().unwrap();
             let callbacks = callbacks.get_mut(&self.id).unwrap();
             if let Some(callback) = callbacks.utterance_begin.as_mut() {
@@ -210,6 +222,18 @@ impl Backend for WinRt {
         Ok(Some(utterance_id))
     }
 
+    fn synthesize(&mut self, text: &str) -> Result<Vec<u8>, Error> {
+        let stream = self.create_synthesis_stream(text)?;
+
+        let size = stream.Size()?;
+        let data_reader = DataReader::CreateDataReader(&stream.GetInputStreamAt(0)?)?;
+        let mut bytes = vec![0; size as usize];
+        data_reader.LoadAsync(size as u32)?;
+        data_reader.ReadBytes(&mut bytes)?;
+
+        Ok(bytes)
+    }
+
     fn stop(&mut self) -> std::result::Result<(), Error> {
         trace!("stop()");
         if !self.is_speaking()? {

src/lib.rs

@@ -162,6 +162,7 @@ unsafe impl Sync for UtteranceId {}
 #[cfg_attr(feature = "serde", derive(serde::Serialize, serde::Deserialize))]
 pub struct Features {
     pub is_speaking: bool,
+    pub synthesize: bool,
     pub pitch: bool,
     pub rate: bool,
     pub stop: bool,
@@ -217,6 +218,7 @@ pub trait Backend: Clone {
     fn id(&self) -> Option<BackendId>;
     fn supported_features(&self) -> Features;
     fn speak(&mut self, text: &str, interrupt: bool) -> Result<Option<UtteranceId>, Error>;
+    fn synthesize(&mut self, text: &str) -> Result<Vec<u8>, Error>;
     fn stop(&mut self) -> Result<(), Error>;
     fn min_rate(&self) -> f32;
     fn max_rate(&self) -> f32;
@@ -374,6 +376,15 @@ impl Tts {
             .speak(text.into().as_str(), interrupt)
     }
 
+    pub fn synthesize<S: Into<String>>(&mut self, text: S) -> Result<Vec<u8>, Error> {
+        let Features { synthesize, .. } = self.supported_features();
+        if synthesize {
+            self.0.write().unwrap().synthesize(text.into().as_str())
+        } else {
+            Err(Error::UnsupportedFeature)
+        }
+    }
+
     /// Stops current speech.
     pub fn stop(&mut self) -> Result<&Self, Error> {
         let Features { stop, .. } = self.supported_features();
@@ -571,7 +582,7 @@ impl Tts {
         if utterance_callbacks {
             let mut callbacks = CALLBACKS.lock().unwrap();
             let id = self.0.read().unwrap().id().unwrap();
-            let mut callbacks = callbacks.get_mut(&id).unwrap();
+            let callbacks = callbacks.get_mut(&id).unwrap();
             callbacks.utterance_begin = callback;
             Ok(())
         } else {
@@ -591,7 +602,7 @@ impl Tts {
         if utterance_callbacks {
             let mut callbacks = CALLBACKS.lock().unwrap();
             let id = self.0.read().unwrap().id().unwrap();
-            let mut callbacks = callbacks.get_mut(&id).unwrap();
+            let callbacks = callbacks.get_mut(&id).unwrap();
             callbacks.utterance_end = callback;
             Ok(())
         } else {
@@ -611,7 +622,7 @@ impl Tts {
         if utterance_callbacks {
             let mut callbacks = CALLBACKS.lock().unwrap();
             let id = self.0.read().unwrap().id().unwrap();
-            let mut callbacks = callbacks.get_mut(&id).unwrap();
+            let callbacks = callbacks.get_mut(&id).unwrap();
             callbacks.utterance_stop = callback;
             Ok(())
         } else {