语音控制

本模块提供机器人语音相关的所有功能，包括语音识别、语音合成、语音控制等。

voice 高层接口（推荐）

底层接口（精细控制）

单例与基础

static Speech.get_instance()

Get the singleton instance of Speech.

Returns:: Singleton instance.
Return type:: Speech

Examples:

speech = Speech.get_instance()

AFE (音频前处理)

Speech.enable_afe_engine(enable, block=True, timeout=5)

Enable or disable Audio Front End (AFE) engine.

Parameters:

enable (bool) – True to enable, False to disable.
timeout (int) – Timeout in seconds (default: 5).

Return type:

EnableServiceResponse

Returns:

EnableAfeEngineResponse.

Examples:

# Enable AFE
enable_afe_engine(True)

# Disable AFE with custom timeout
enable_afe_engine(False, timeout=10)

Parameters:: block (bool) –

Speech.enable_aec(enable, block=True, timeout=5)

Enable or disable Acoustic Echo Cancellation (AEC).

Parameters:

enable (bool) – True to enable, False to disable.
timeout (int) – Timeout in seconds (default: 5).

Return type:

EnableServiceResponse

Returns:

EnableAecResponse.

Examples:

enable_aec(True)
enable_aec(False, timeout=8)

Parameters:: block (bool) –

Speech.enable_speech_enhance(enable, block=True, timeout=5)

Enable or disable speech enhancement.

Parameters:

enable (bool) – True to enable, False to disable.
timeout (int) – Timeout in seconds (default: 5).

Return type:

EnableServiceResponse

Returns:

EnableSpeechEnhanceResponse.

Examples:

enable_speech_enhance(True)
enable_speech_enhance(False)

Parameters:: block (bool) –

Speech.set_bypass(bypass, block=True, timeout=5)

Set AFE bypass mode.

Parameters:

bypass (bool) – True to bypass AFE processing, False to use AFE.
timeout (int) – Timeout in seconds (default: 5).

Return type:

EnableServiceResponse

Returns:

SetBypassResponse.

Examples:

set_bypass(True)
set_bypass(False, timeout=10)

Parameters:: block (bool) –

Speech.mute_clean_mic(mute, block=True, timeout=5)

Mute or unmute clean microphone.

Parameters:

mute (bool) – True to mute, False to unmute.
timeout (int) – Timeout in seconds (default: 5).

Return type:

EnableServiceResponse

Returns:

MuteCleanMicResponse.

Examples:

mute_clean_mic(True)
mute_clean_mic(False, timeout=6)

Parameters:: block (bool) –

ASR (语音识别)

Speech.enable_asr(enable, block=True, timeout=30)

Enable or disable Automatic Speech Recognition (ASR).

Parameters:

enable (bool) – True to enable, False to disable.
timeout (int) – Timeout in seconds (default: 30).

Return type:

EnableServiceResponse

Returns:

EnableAsrResponse.

Examples:

enable_asr(True)
enable_asr(False, timeout=8)

Parameters:: block (bool) –

Speech.switch_asr_device(device_name, timeout=5)

Switch ASR input device.

Parameters:

device_name (str) – Device name for ASR input.
timeout (int) – Timeout in seconds (default: 5).

Return type:

EnableServiceResponse

Returns:

SwitchAsrDeviceResponse.

Examples:

switch_asr_device("default")
switch_asr_device("mic_array", timeout=10)

KWS (关键词唤醒)

Speech.enable_hexaware(enable, block=True, timeout=5)

Enable or disable Hexaware wake word detection.

Parameters:

enable (bool) – True to enable, False to disable.
timeout (int) – Timeout in seconds (default: 5).

Return type:

EnableServiceResponse

Returns:

EnableHexawareResponse.

Examples:

enable_hexaware(True)
enable_hexaware(False, timeout=8)

Parameters:: block (bool) –

Speech.enable_kws(enable, block=True, timeout=5)

Enable or disable keyword spotting (KWS).

Parameters:

enable (bool) – True to enable, False to disable.
timeout (int) – Timeout in seconds (default: 5).

Return type:

EnableServiceResponse

Returns:

EnableKwsResponse.

Examples:

enable_kws(True)
enable_kws(False, timeout=8)

Parameters:: block (bool) –

TTS (语音合成)

Speech.enable_tts(enable, block=True, timeout=30)

Enable or disable Text-to-Speech (TTS) engine.

Parameters:

enable (bool) – True to enable, False to disable.
timeout (int) – Timeout in seconds (default: 30).

Return type:

EnableServiceResponse

Returns:

EnableTtsResponse.

Examples:

enable_tts(True)
enable_tts(False, timeout=8)

Parameters:: block (bool) –

Speech.stop_playing_tts(enable, block=True, timeout=5)

Stop playing current TTS audio.

Parameters:

enable (bool) – True to stop, False to ignore.
timeout (int) – Timeout in seconds (default: 5).

Return type:

EnableServiceResponse

Returns:

StopPlayingTtsResponse.

Examples:

stop_playing_tts(True)
stop_playing_tts(True, timeout=8)

Parameters:: block (bool) –

Speech.switch_tts_device(device_name, timeout=5)

Switch TTS output device.

Parameters:

device_name (str) – Device name for TTS output.
timeout (int) – Timeout in seconds (default: 5).

Return type:

EnableServiceResponse

Returns:

SwitchTtsDeviceResponse.

Examples:

switch_tts_device("default")
switch_tts_device("speaker", timeout=10)

Speech.set_volume(volume, timeout=5)

Set TTS playback volume.

Parameters:

volume (float) – Volume in the range [0.0, 1.0].
timeout (int) – Timeout in seconds (default: 5).

Return type:

EnableServiceResponse

Returns:

SetVolumeResponse.

Examples:

set_volume(0.5)
set_volume(1.0, timeout=10)

Speech.generate_audio(text, sid=0, speed=1.0, block=True, timeout=60, complete_cb=None, feedback_cb=None)

Generate and play audio from text.

Parameters:

text (str) – Text to convert to speech.
sid (int) – Speaker ID (default: 0).
speed (float) – Speech speed (default: 1.0).
block (bool) – Whether to block until completion (default: True).
timeout (int) – Timeout in seconds (default: 60).
complete_cb (callable) – Callback for completion (default: None).
feedback_cb (callable) – Callback for progress updates (default: None).

Return type:

GenerateAudioResponse

Returns:

GenerateAudioResponse.

Examples:

# Simple usage
generate_audio("Hello, world!")

# Customize speaker and speed
generate_audio("Hello", sid=1, speed=1.2)

# Non-blocking
generate_audio(text="Hello", block=False)

# With callbacks
def on_complete(response):
    print(f"Done: {response.result.message}")

def on_progress(event):
    print(f"Progress: {event.progress * 100:.1f}%")

generate_audio("Hello", complete_cb=on_complete, feedback_cb=on_progress)

Speech.cancel_generate_audio(timeout=5)

Cancel ongoing audio generation.

Parameters:: timeout (int) – Timeout in seconds (default: 5).
Return type:: EnableServiceResponse
Returns:: CancelGenerateAudioResponse.

Examples:

# Cancel current audio generation
cancel_generate_audio()

# Cancel with custom timeout
cancel_generate_audio(timeout=8)

事件回调

Speech.register_asr_callback(callback=None)

Register callback for ASR (speech recognition) results.

Parameters:: callback (callable) – Callback function that receives AsrResultEvent. Pass None to unregister.
Return type:: None
Returns:: None.

Examples:

def on_asr(event):
    print(f"Recognized: {event.text}")

register_asr_callback(on_asr)
# Unregister
register_asr_callback(None)

Speech.register_hexaware_callback(callback=None)

Register callback for wake word detection.

Parameters:: callback (callable) – Callback function that receives HexawareAwakenEvent. Pass None to unregister.
Return type:: None
Returns:: None.

Examples:

def on_wake_word(event):
    print(f"Wake word detected: {event.keyword}")

register_hexaware_callback(on_wake_word)
register_hexaware_callback(None)

Speech.register_kws_message_callback(callback=None)

Register callback for keyword spotting messages.

Parameters:: callback (callable) – Callback function that receives KwsMessageEvent. Pass None to unregister.
Return type:: None
Returns:: None.

Examples:

def on_kws_message(event):
    print(f"KWS message: {event.message}")

register_kws_message_callback(on_kws_message)
register_kws_message_callback(None)

Speech.register_nlu_format_callback(callback=None)

Register callback for NLU format messages.

Parameters:: callback (callable) – Callback function that receives NluFormatEvent. Pass None to unregister.
Return type:: None
Returns:: None.

Examples:

def on_nlu_format(event):
    print(f"Input: {event.input_text}, Format: {event.format_text}")

register_nlu_format_callback(on_nlu_format)
register_nlu_format_callback(None)

Speech.register_generate_audio_callbacks(complete_cb, feedback_cb)

Register callbacks for audio generation progress and completion.

Parameters:

complete_cb (callable) – Callback for completion, receives GenerateAudioResponse.
feedback_cb (callable) – Callback for progress updates, receives GenerateAudioProgressEvent.

Return type:

None

Returns:

None.

Examples:

def on_complete(response):
    print(f"Audio generation complete: {response.result.message}")

def on_progress(event):
    print(f"Progress: {event.progress * 100:.1f}%")

register_generate_audio_callbacks(on_complete, on_progress)