SOTA Open Source TTS
Streaming Agent (#659)
spicysama672000d59467
diff --git a/tools/e2e_webui.py b/tools/e2e_webui.pyindex 68cfe1d..37474fb 100644--- a/tools/e2e_webui.py+++ b/tools/e2e_webui.py@@ -1,4 +1,6 @@+import ioimport re+import waveimport gradio as grimport numpy as np@@ -7,6 +9,19 @@ from .fish_e2e import FishE2EAgent, FishE2EEventTypefrom .schema import ServeMessage, ServeTextPart, ServeVQPart+def wav_chunk_header(sample_rate=44100, bit_depth=16, channels=1):+ buffer = io.BytesIO()++ with wave.open(buffer, "wb") as wav_file:+ wav_file.setnchannels(channels)+ wav_file.setsampwidth(bit_depth // 8)+ wav_file.setframerate(sample_rate)++ wav_header_bytes = buffer.getvalue()+ buffer.close()+ return wav_header_bytes++class ChatState:def __init__(self):self.conversation = []@@ -62,7 +77,7 @@ async def process_audio_input(if isinstance(sys_audio_input, tuple):sr, sys_audio_data = sys_audio_input- elif text_input:+ else:sr = 44100sys_audio_data = None@@ -95,21 +110,13 @@ async def process_audio_input(if event.type == FishE2EEventType.USER_CODES:append_to_chat_ctx(ServeVQPart(codes=event.vq_codes), role="user")elif event.type == FishE2EEventType.SPEECH_SEGMENT:- result_audio += event.frame.data- np_audio = np.frombuffer(result_audio, dtype=np.int16)append_to_chat_ctx(ServeVQPart(codes=event.vq_codes))-- yield state.get_history(), (44100, np_audio), None, None+ yield state.get_history(), wav_chunk_header() + event.frame.data, None, Noneelif event.type == FishE2EEventType.TEXT_SEGMENT:append_to_chat_ctx(ServeTextPart(text=event.text))- if result_audio:- np_audio = np.frombuffer(result_audio, dtype=np.int16)- yield state.get_history(), (44100, np_audio), None, None- else:- yield state.get_history(), None, None, None+ yield state.get_history(), None, None, None- np_audio = np.frombuffer(result_audio, dtype=np.int16)- yield state.get_history(), (44100, np_audio), None, None+ yield state.get_history(), None, None, Noneasync def process_text_input(@@ -179,7 +186,12 @@ def create_demo():text_input = gr.Textbox(label="Or type your message", type="text")- output_audio = gr.Audio(label="Assistant's Voice", type="numpy")+ output_audio = gr.Audio(+ label="Assistant's Voice",+ streaming=True,+ autoplay=True,+ interactive=False,+ )send_button = gr.Button("Send", variant="primary")clear_button = gr.Button("Clear")