Add voice speed.

Author: p-groarke

libinclude/wsay/wsay.hpp

@@ -49,10 +49,11 @@ struct voice {
 	// Stop writing wave file.
 	void stop_file_output();
 
-
 	// Sets the voice volume, from 0 to 100.
 	void set_volume(uint16_t volume);
 
+	// Sets the voice speed, from 0 to 100.
+	void set_speed(uint16_t speed);
 
 	// Speaks the sentence using selected voice to playback outputs and file.
 	// Blocking.

libsrc/wsay.cpp

@@ -134,19 +134,20 @@ bool parse_text_file(
 
 struct voice_data {
 	// Creates default voice.
-	voice_data(ISpObjectToken* selected_voice, uint16_t vol) {
+	voice_data(ISpObjectToken* selected_voice, uint16_t vol, int speed) {
 		if (!SUCCEEDED(voice_ptr.CoCreateInstance(CLSID_SpVoice))) {
 			throw std::runtime_error{ "Couldn't initialize voice." };
 		}
 
 		voice_ptr->SetVoice(selected_voice);
 		voice_ptr->SetVolume(vol);
+		voice_ptr->SetRate(speed);
 	}
 
 	// Creates file out voice.
 	voice_data(ISpObjectToken* selected_voice,
-			const std::filesystem::path& filepath, uint16_t vol)
-			: voice_data(selected_voice, vol) {
+			const std::filesystem::path& filepath, uint16_t vol, int speed)
+			: voice_data(selected_voice, vol, speed) {
 		path = filepath;
 
 		CComPtr<ISpStream> cpStream;
@@ -171,8 +172,8 @@ struct voice_data {
 	}
 
 	voice_data(ISpObjectToken* selected_voice, size_t device_idx,
-			ISpObjectToken* device, uint16_t vol)
-			: voice_data(selected_voice, vol) {
+			ISpObjectToken* device, uint16_t vol, int speed)
+			: voice_data(selected_voice, vol, speed) {
 
 		device_playback_idx = device_idx;
 		voice_ptr->SetOutput(device, TRUE);
@@ -203,7 +204,7 @@ struct voice_impl {
 
 		available_devices = get_playback_devices();
 
-		voices.push_back(voice_data{ nullptr, 100u });
+		voices.push_back(voice_data{ nullptr, 100u, 0 });
 	}
 
 	template <class Func>
@@ -229,6 +230,7 @@ struct voice_impl {
 	std::vector<voice_data> voices;
 	size_t selected_voice = 0;
 	uint16_t volume = 100;
+	int speed = 0;
 };
 
 
@@ -321,6 +323,7 @@ void voice::enable_device_playback(size_t device_idx) {
 			device_idx,
 			_impl->available_devices[device_idx].second,
 			_impl->volume,
+			_impl->speed,
 	});
 }
 
@@ -354,6 +357,7 @@ void voice::start_file_output(const std::filesystem::path& path) {
 			_impl->available_voices[_impl->selected_voice].second,
 			path,
 			_impl->volume,
+			_impl->speed,
 	});
 }
 
@@ -381,6 +385,17 @@ void voice::set_volume(uint16_t volume) {
 	}
 }
 
+void voice::set_speed(uint16_t speed) {
+	speed = std::clamp(speed, uint16_t(0), uint16_t(100));
+	double temp = (speed / 100.0) * 20.0;
+
+	_impl->speed = int(temp) - 10;
+
+	for (voice_data& v : _impl->voices) {
+		v.voice_ptr->SetRate(_impl->speed);
+	}
+}
+
 void voice::speak(const std::wstring& sentence) {
 	_impl->execute([&](CComPtr<ISpVoice>& voice) {
 		voice->Speak(sentence.c_str(), SPF_DEFAULT | SPF_ASYNC, nullptr);

readme.md

@@ -17,6 +17,7 @@ Simple command line text-to-speech with easy file output, voice selection and mo
   - Note : Window's legacy command prompt has some of it's own issues with utf8.
   - If you need full unicode support, best to use the new [Windows Terminal](https://aka.ms/terminal).
 - Supports utf8, utf16le and utf16be text files.
+- Volume and speed options.
 
 
 ## Install Instructions
@@ -84,11 +85,17 @@ wsay "I conquer all devices" -p all
 # You can set the voice volume, from 0 to 100.
 wsay "Softly speaking" --volume 25
 
+# You can set the voice speed, from 0 to 100. 50 is the default speed.
+wsay "Quickly speaking" --speed 75
+
 # Here, we are using voice 6, reading text from a file and outputting to 'output.wav'.
 wsay -v 6 -i mix_and_match_options.txt -o output.wav
 
 # Ouput to multiple devices using interactive mode with voice 5.
 wsay -v 5 -I -p 1 2
+
+# Speak slowly and quietly, on all devices, using voice 7 and save to wav file.
+wsay "Multiple options example." -v 7 -p all -s 25 -V 25 -o
 ```
 
 
@@ -110,6 +117,7 @@ Options:
                                    You can provide more than 1 playback device, seperate the numbers with spaces. You
                                    can also mix output to file + playback.
                                    Use 'all' to select all devices.
+ -s, --speed <value>               Sets the voice speed, from 0 to 100. 50 is the default speed.
  -v, --voice <value>               Choose a different voice. Use the voice number printed using --list_voices.
  -V, --volume <value>              Sets the voice volume, from 0 to 100.
  -h, --help                        Print this help

src_cmd/main.cpp

@@ -171,6 +171,16 @@ int wmain(int argc, wchar_t** argv, wchar_t**) {
 			},
 			L"Sets the voice volume, from 0 to 100.", L'V');
 
+	opt.add_required_arg_option(
+			L"speed",
+			[&](std::wstring&& f) {
+				uint16_t speed = uint16_t(std::stoul(f));
+				voice.set_speed(speed);
+				return true;
+			},
+			L"Sets the voice speed, from 0 to 100. 50 is the default speed.",
+			L's');
+
 	std::wstring help_outro = L"wsay\nversion ";
 	help_outro += WSAY_VERSION;
 	help_outro += L"\nhttps://github.com/p-groarke/wsay/releases\n";