diff --git a/include/tts/TtsEngine.h b/include/tts/TtsEngine.h
new file mode 100644
index 0000000..21cb73b
--- /dev/null
+++ b/include/tts/TtsEngine.h
@@ -0,0 +1,237 @@
+/*
+ * Copyright (C) 2009 Google Inc.
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *      http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+#include <media/AudioSystem.h>
+
+// This header defines the interface used by the Android platform
+// to access Text-To-Speech functionality in shared libraries that implement
+// speech synthesis and the management of resources associated with the
+// synthesis.
+// An example of the implementation of this interface can be found in
+// FIXME: add path+name to implementation of default TTS engine
+// Libraries implementing this interface are used in:
+//  frameworks/base/tts/jni/android_tts_SpeechSynthesis.cpp
+
+namespace android {
+
+enum tts_synth_status {
+    TTS_SYNTH_DONE              = 0,
+    TTS_SYNTH_PENDING           = 1
+};
+
+enum tts_callback_status {
+    TTS_CALLBACK_HALT           = 0,
+    TTS_CALLBACK_CONTINUE       = 1
+};
+
+// The callback is used by the implementation of this interface to notify its
+// client, the Android TTS service, that the last requested synthesis has been
+// completed. // TODO reword
+// The callback for synthesis completed takes:
+// @param [inout] void *&       - The userdata pointer set in the original
+//                                 synth call
+// @param [in]    uint32_t      - Track sampling rate in Hz
+// @param [in]    audio_format  - The AudioSystem::audio_format enum
+// @param [in]    int           - The number of channels
+// @param [inout] int8_t *&     - A buffer of audio data only valid during the
+//                                execution of the callback
+// @param [inout] size_t  &     - The size of the buffer
+// @param [in] tts_synth_status - indicate whether the synthesis is done, or
+//                                 if more data is to be synthesized.
+// @return TTS_CALLBACK_HALT to indicate the synthesis must stop,
+//         TTS_CALLBACK_CONTINUE to indicate the synthesis must continue if
+//            there is more data to produce.
+typedef tts_callback_status (synthDoneCB_t)(void *&, uint32_t,
+        AudioSystem::audio_format, int, int8_t *&, size_t&, tts_synth_status);
+
+class TtsEngine;
+extern "C" TtsEngine* getTtsEngine();
+
+enum tts_result {
+    TTS_SUCCESS                 = 0,
+    TTS_FAILURE                 = -1,
+    TTS_FEATURE_UNSUPPORTED     = -2,
+    TTS_VALUE_INVALID           = -3,
+    TTS_PROPERTY_UNSUPPORTED    = -4,
+    TTS_PROPERTY_SIZE_TOO_SMALL = -5,
+    TTS_MISSING_RESOURCES       = -6
+};
+
+enum tts_support_result {
+    TTS_LANG_COUNTRY_VAR_AVAILABLE = 2,
+    TTS_LANG_COUNTRY_AVAILABLE = 1,
+    TTS_LANG_AVAILABLE = 0,
+    TTS_LANG_MISSING_DATA = -1,
+    TTS_LANG_NOT_SUPPORTED = -2
+};
+
+class TtsEngine
+{
+public:
+    // Initialize the TTS engine and returns whether initialization succeeded.
+    // @param synthDoneCBPtr synthesis callback function pointer
+    // @return TTS_SUCCESS, or TTS_FAILURE
+    virtual tts_result init(synthDoneCB_t synthDoneCBPtr);
+
+    // Shut down the TTS engine and releases all associated resources.
+    // @return TTS_SUCCESS, or TTS_FAILURE
+    virtual tts_result shutdown();
+
+    // Interrupt synthesis and flushes any synthesized data that hasn't been
+    // output yet. This will block until callbacks underway are completed.
+    // @return TTS_SUCCESS, or TTS_FAILURE
+    virtual tts_result stop();
+
+    // Returns the level of support for the language, country and variant.
+    // @return TTS_LANG_COUNTRY_VAR_AVAILABLE if the language, country and variant are supported,
+    //            and the corresponding resources are correctly installed
+    //         TTS_LANG_COUNTRY_AVAILABLE if the language and country are supported and the
+    //             corresponding resources are correctly installed, but there is no match for
+    //             the specified variant
+    //         TTS_LANG_AVAILABLE if the language is supported and the
+    //             corresponding resources are correctly installed, but there is no match for
+    //             the specified country and variant
+    //         TTS_LANG_MISSING_DATA if the required resources to provide any level of support
+    //             for the language are not correctly installed
+    //         TTS_LANG_NOT_SUPPORTED if the language is not supported by the TTS engine.
+    virtual tts_support_result isLanguageAvailable(const char *lang, const char *country,
+            const char *variant);
+
+    // Load the resources associated with the specified language. The loaded
+    // language will only be used once a call to setLanguage() with the same
+    // language value is issued. Language and country values are coded according to the ISO three
+    // letter codes for languages and countries, as can be retrieved from a java.util.Locale
+    // instance. The variant value is encoded as the variant string retrieved from a
+    // java.util.Locale instance built with that variant data.
+    // @param lang pointer to the ISO three letter code for the language
+    // @param country pointer to the ISO three letter code for the country
+    // @param variant pointer to the variant code
+    // @return TTS_SUCCESS, or TTS_FAILURE
+    virtual tts_result loadLanguage(const char *lang, const char *country, const char *variant);
+    
+    // Load the resources associated with the specified language, country and Locale variant.
+    // The loaded language will only be used once a call to setLanguageFromLocale() with the same
+    // language value is issued. Language and country values are coded according to the ISO three
+    // letter codes for languages and countries, as can be retrieved from a java.util.Locale
+    // instance. The variant value is encoded as the variant string retrieved from a
+    // java.util.Locale instance built with that variant data.
+    // @param lang pointer to the ISO three letter code for the language
+    // @param country pointer to the ISO three letter code for the country
+    // @param variant pointer to the variant code
+    // @return TTS_SUCCESS, or TTS_FAILURE
+    virtual tts_result setLanguage(const char *lang, const char *country, const char *variant);
+
+    // Retrieve the currently set language, country and variant, or empty strings if none of
+    // parameters have been set. Language and country are represented by their 3-letter ISO code
+    // @param[out]   pointer to the retrieved 3-letter code language value
+    // @param[out]   pointer to the retrieved 3-letter code country value
+    // @param[out]   pointer to the retrieved variant value
+    // @return TTS_SUCCESS, or TTS_FAILURE
+    virtual tts_result getLanguage(char *language, char *country, char *variant);
+
+    // Notifies the engine what audio parameters should be used for the synthesis.
+    // This is meant to be used as a hint, the engine implementation will set the output values
+    // to those of the synthesis format, based on a given hint.
+    // @param[inout] encoding in: the desired audio sample format
+    //                         out: the format used by the TTS engine
+    // @param[inout] rate in: the desired audio sample rate
+    //                         out: the sample rate used by the TTS engine
+    // @param[inout] channels in: the desired number of audio channels
+    //                         out: the number of channels used by the TTS engine
+    // @return TTS_SUCCESS, or TTS_FAILURE
+    virtual tts_result setAudioFormat(AudioSystem::audio_format& encoding, uint32_t& rate,
+            int& channels);
+
+    // Set a property for the the TTS engine
+    // "size" is the maximum size of "value" for properties "property"
+    // @param property pointer to the property name
+    // @param value    pointer to the property value
+    // @param size     maximum size required to store this type of property
+    // @return         TTS_PROPERTY_UNSUPPORTED, or TTS_SUCCESS, or TTS_FAILURE,
+    //                  or TTS_VALUE_INVALID
+    virtual tts_result setProperty(const char *property, const char *value,
+            const size_t size);
+
+    // Retrieve a property from the TTS engine
+    // @param        property pointer to the property name
+    // @param[out]   value    pointer to the retrieved language value
+    // @param[inout] iosize   in: stores the size available to store the
+    //                          property value.
+    //                        out: stores the size required to hold the language
+    //                          value if getLanguage() returned
+    //                          TTS_PROPERTY_SIZE_TOO_SMALL, unchanged otherwise
+    // @return TTS_PROPERTY_UNSUPPORTED, or TTS_SUCCESS,
+    //         or TTS_PROPERTY_SIZE_TOO_SMALL
+    virtual tts_result getProperty(const char *property, char *value,
+            size_t *iosize);
+
+    // Synthesize the text.
+    // As the synthesis is performed, the engine invokes the callback to notify
+    // the TTS framework that it has filled the given buffer, and indicates how
+    // many bytes it wrote. The callback is called repeatedly until the engine
+    // has generated all the audio data corresponding to the text.
+    // Note about the format of the input: the text parameter may use the
+    // following elements
+    // and their respective attributes as defined in the SSML 1.0 specification:
+    //    * lang
+    //    * say-as:
+    //          o interpret-as
+    //    * phoneme
+    //    * voice:
+    //          o gender,
+    //          o age,
+    //          o variant,
+    //          o name
+    //    * emphasis
+    //    * break:
+    //          o strength,
+    //          o time
+    //    * prosody:
+    //          o pitch,
+    //          o contour,
+    //          o range,
+    //          o rate,
+    //          o duration,
+    //          o volume
+    //    * mark
+    // Differences between this text format and SSML are:
+    //    * full SSML documents are not supported
+    //    * namespaces are not supported
+    // Text is coded in UTF-8.
+    // @param text      the UTF-8 text to synthesize
+    // @param userdata  pointer to be returned when the call is invoked
+    // @param buffer    the location where the synthesized data must be written
+    // @param bufferSize the number of bytes that can be written in buffer
+    // @return          TTS_SUCCESS or TTS_FAILURE
+    virtual tts_result synthesizeText(const char *text, int8_t *buffer,
+            size_t bufferSize, void *userdata);
+
+    // Synthesize IPA text.
+    // As the synthesis is performed, the engine invokes the callback to notify
+    // the TTS framework that it has filled the given buffer, and indicates how
+    // many bytes it wrote. The callback is called repeatedly until the engine
+    // has generated all the audio data corresponding to the IPA data.
+    // @param ipa      the IPA data to synthesize
+    // @param userdata  pointer to be returned when the call is invoked
+    // @param buffer    the location where the synthesized data must be written
+    // @param bufferSize the number of bytes that can be written in buffer
+    // @return TTS_FEATURE_UNSUPPORTED if IPA is not supported,
+    //         otherwise TTS_SUCCESS or TTS_FAILURE
+    virtual tts_result synthesizeIpa(const char *ipa, int8_t *buffer,
+            size_t bufferSize, void *userdata);
+};
+
+} // namespace android
+