deepgeek/gradio-webrtc
2
0
Fork 0
mirror of https://github.com/HumanAIGC-Engineering/gradio-webrtc.git synced 2026-02-05 09:59:22 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix Websocket Client Processing (#286)

Browse Source 
* Fix

* Add code
This commit is contained in:
Freddy Boulton
2025-04-17 12:21:13 -04:00
committed by GitHub
parent c9bca428af
commit a68023101d
4 changed files with 52 additions and 36 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
backend/fastrtc/websocket.py
View File

@@ -43,7 +43,9 @@ def convert_to_mulaw(
audio_data = audio_to_float32(audio_data)
if original_rate != target_rate:
audio_data = librosa.resample(audio_data, orig_sr=original_rate, target_sr=8000)
audio_data = librosa.resample(
audio_data, orig_sr=original_rate, target_sr=target_rate
)
audio_data = audio_to_int16(audio_data)
@@ -128,7 +130,10 @@ class WebSocketHandler:
audioop.ulaw2lin(audio_payload, 2), dtype=np.int16
)
if self.stream_handler.input_sample_rate != 8000:
if (
self.stream_handler.phone_mode
and self.stream_handler.input_sample_rate != 8000
):
audio_array = audio_to_float32(audio_array)
audio_array = librosa.resample(
audio_array,
@@ -269,14 +274,15 @@ class WebSocketHandler:
if not isinstance(frame, tuple):
continue
target_rate = (
self.stream_handler.output_sample_rate
if not self.stream_handler.phone_mode
else 8000
8_000
if self.stream_handler.phone_mode
else self.stream_handler.output_sample_rate
)
mulaw_audio = convert_to_mulaw(
frame[1], frame[0], target_rate=target_rate
frame[1],
frame[0],
target_rate=target_rate,
)
audio_payload = base64.b64encode(mulaw_audio).decode("utf-8")