uniMRCP + UCCX + Azure (local containers), RTP arrives but ASR returns no-input; guidance requested

52 views

Skip to first unread message

Daniel Friedman

unread,

Sep 7, 2025, 9:30:27 AMSep 7

to UniMRCP

Hi UniMRCP Support Team,

We’re troubleshooting an integration where UCCX talks to UniMRCP (masquerading as LumenVox)

and UniMRCP fans out to local Azure Speech containers (STT/TTS).

RTP clearly reaches UniMRCP, but recognition consistently ends with no-input.

Details below, plus a call log attached.

Environment / versions

UniMRCP Server: 1.7.0
SofiaSIP: 1.12.11-239-g54ef3e2
APR: 1.5.2
Azure STT plugin (umsazuresr): 1.32.0
Azure TTS plugin (umsazuress): 1.25.1
Licenses: SR 2 ch, SS 4 ch
OS: Debian-based VM; packages upgraded via apt-get to latest available

Architecture

UCCX → UniMRCP (we present as “LumenVox Media Server” because UCCX doesn’t support UniMRCP server directly).
UniMRCP → Azure local containers
- STT endpoint(s): http://localhost:5001 / http://localhost:5002
- TTS endpoint: https://eastus.tts.speech.microsoft.com/cognitiveservices/v1 (for now until we will install a local container)

What works

SIP/MRCPv2 signaling is clean. Channel comes up, grammars are defined, and RECOGNIZE requests are accepted.
RTP session opens: UniMRCP answers with PCMU/8000 (recvonly).
We captured the RTP stream at the UniMRCP host and can play it in Wireshark, you can clearly hear “yes” / “no”.
The Azure STT WebSocket session is established and upgraded (101 Switching Protocols).

Symptoms

Despite RTP arriving, the UniMRCP log shows:
- Input Complete [success] size=0 bytes, dur=0 ms
- Followed by RECOGNITION-COMPLETE ... Completion-Cause: 002 no-input-timeout
Periodic warnings appear when MRCP messages are received (from UCCX):
- WARN Cannot read the first field in start-line
  (Messages still seem to be processed, grammar loads, recognize starts, so unsure if this is benign parsing noise or relevant.)

Evidence from logs (high level)

RTP receive: e.g., Open RTP Receiver 10.80.64.150:4002 <- 10.80.60.105:25308
Media path: Source->[PCMU/8000/1]->Decoder->[LPCM/8000/1]->Bridge->[LPCM/8000/1]->Sink
Files opened (for debugging): both 8 kHz and 16 kHz WAV writers open, and a resampler to 16 kHz is created for the Azure STT container.
VAD: start-of-input=internal, noinput-timeout=10000 ms, VAD mode 0; yet recorded size remains 0.

Config highlights (current)

SIP transport: TCP 5060.
MRCPv2: passive setup on server, port 1544.
RTP: PCMU/8000 and telephone-event advertised; jitter buffer enabled; ptime 20 ms.
Profiles: speechrecog → Azure-SR-1 (umsazuresr 1.32.0); speechsynth → Azure-SS-1.

What we’ve already tried

Verified audio path with PCAP (playable, intelligible).
Reduced advertised codecs to PCMU + DTMF.
Cleaned up unimrcpserver.xml to remove non-schema elements; resolved earlier config warnings.
Upgraded UniMRCP modules (now on umsazuresr 1.32.0; TTS remains 1.25.1).

Questions for your guidance

VAD / audio handoff: Given that RTP is received and decodes to LPCM, what would cause size=0 bytes on the recognizer side? Any known issues in umsazuresr 1.32.0 with μ-law 8 kHz input, internal VAD, or resampling to 16 kHz before sending to Azure?
MRCP parse warnings: Is Cannot read the first field in start-line expected with UCCX’s MRCPv2 framing, or does it indicate a line-ending/BOM/framing quirk that could influence VAD or timing?
Best-practice knobs:
- Should we set start-of-input = external and rely solely on UCCX’s START-INPUT-TIMERS, or keep internal?
- Any recommended VAD parameters (e.g., thresholds, leading/trailing silence) for 8 kHz μ-law from Cisco gateways/UCCX?
- Any reason to prefer InputFormat over AudioFormat for STT in 1.32.0?
Codec/ptime: Are there known corner cases with UCCX offering sendonly PCMU and UniMRCP answering recvonly that would require adjusting ptime, jitter buffer, or SOA settings?

Attachments

uniMRCP_Log_9-May-2025.txt (full call flow showing RTP open, RECOGNIZE, VAD/no-input)
PCAP of the call (available on request)

Any pointers, known issues, or recommended config tweaks would be greatly appreciated.

We’re happy to provide additional traces/configs as needed.

Thank you,

Dani

Daniel Friedman

unread,

Sep 7, 2025, 9:34:26 AMSep 7

to UniMRCP

here are the logs:

2025-09-07 15:35:05:551373 [INFO] Receive SIP Event [nua_i_invite] Status 100 Trying [SIP-Agent-1]
2025-09-07 15:35:05:551386 [INFO] Receive SIP Event [nua_i_state] Status 100 Trying [SIP-Agent-1]
2025-09-07 15:35:05:551390 [NOTICE] SIP Call State [received]
2025-09-07 15:35:05:551400 [INFO] Create Session 0x7f6c84002808 <new> [uni2]
2025-09-07 15:35:05:551405 [INFO] Remote SDP 0x7f6c84002808 <new>
v=0
o=- 123 456 IN IP4 10.80.60.105
s=Unified Contact Center Express 12.5.1.11003-511
c=IN IP4 10.80.60.105
t=0 0
m=audio 25448 RTP/AVP 0 96
a=rtpmap:0 pcmu/8000
a=rtpmap:96 telephone-event/8000
a=fmtp:96 0-15
a=sendonly
a=mid:1
m=application 9 TCP/MRCPv2 19
a=setup:active
a=connection:new
a=fmtp:96 0-15
a=resource:speechrecog
a=cmid:1

2025-09-07 15:35:05:553403 [NOTICE] Add Session <f2981b70f54f4ebe>
2025-09-07 15:35:05:553426 [INFO] Receive Offer 0x7f6c84002808 <f2981b70f54f4ebe> [c:1 a:1 v:0]
2025-09-07 15:35:05:553433 [INFO] Found MRCP Engine [Azure-SR-1] for Resource [speechrecog] 0x7f6c84002808 <f2981b70f54f4ebe>
2025-09-07 15:35:05:553472 [INFO] Add Pending Control Channel <f2981b70f54f4ebe@speechrecog> [1]
2025-09-07 15:35:05:557305 [INFO] Enable RTP Session 10.80.64.150:4002
2025-09-07 15:35:05:557324 [INFO] Open RTP Receiver 10.80.64.150:4002 <- 10.80.60.105:25448 playout [50 ms] bounds [0 - 200 ms] adaptive [1] skew detection [1]
2025-09-07 15:35:05:557336 [INFO] Media Path 0x7f6c84002808 Source->[PCMU/8000/1]->Decoder->[LPCM/8000/1]->Bridge->[LPCM/8000/1]->Sink
2025-09-07 15:35:05:557405 [INFO] Open <f2981b70f54f4ebe@azuresr>
2025-09-07 15:35:05:557442 [NOTICE] AZURESR Usage: 1/1/2
2025-09-07 15:35:05:557460 [INFO] Send Answer 0x7f6c84002808 <f2981b70f54f4ebe> [c:1 a:1 v:0] Status OK
2025-09-07 15:35:05:557463 [NOTICE] Usage [umsazuresr] min [0] cur [1] max [1]
2025-09-07 15:35:05:557487 [INFO] Local SDP 0x7f6c84002808 <f2981b70f54f4ebe>
v=0
o=LumenVoxMediaServer 0 0 IN IP4 10.80.64.150
s=-
c=IN IP4 10.80.64.150
t=0 0
m=audio 4002 RTP/AVP 0 101
a=rtpmap:0 PCMU/8000
a=rtpmap:101 telephone-event/8000
a=fmtp:101 0-15
a=recvonly
a=mid:1
m=application 1544 TCP/MRCPv2 1
a=setup:passive
a=connection:new
a=channel:f2981b70f54f4ebe@speechrecog
a=cmid:1

2025-09-07 15:35:05:558069 [INFO] Receive SIP Event [nua_i_state] Status 200 OK [SIP-Agent-1]
2025-09-07 15:35:05:558078 [NOTICE] SIP Call State 0x7f6c84002808 [completed]
2025-09-07 15:35:05:572983 [INFO] Receive SIP Event [nua_i_ack] Status 200 OK [SIP-Agent-1]
2025-09-07 15:35:05:572999 [INFO] Receive SIP Event [nua_i_state] Status 200 OK [SIP-Agent-1]
2025-09-07 15:35:05:573004 [NOTICE] SIP Call State 0x7f6c84002808 [ready]
2025-09-07 15:35:05:573007 [INFO] Receive SIP Event [nua_i_active] Status 200 Call active [SIP-Agent-1]
2025-09-07 15:35:05:573634 [NOTICE] Accepted TCP/MRCPv2 Connection 10.80.64.150:1544 <-> 10.80.60.105:54388
2025-09-07 15:35:05:573880 [INFO] Receive MRCPv2 Data 10.80.64.150:1544 <-> 10.80.60.105:54388 [763 bytes]
MRCP/2.0 761 DEFINE-GRAMMAR 100
Channel-Identifier: f2981b70f54f4ebe@speechrecog
Content-Type:application/srgs+xml
Speech-Language:en-US
Content-Id:SimpleRe...@DialogObject.grammar
Content-Length: 544

<?xml version="1.0" encoding="UTF-8"?>
<grammar root="main" tag-format="swi-semantics/1.0" version="1.0" xml:lang="en-US" xmlns="http://www.w3.org/2001/06/grammar">
<rule id="main" scope="public">
<one-of>
<item>
<ruleref uri="#MI_1_elem0_rule"/>
<tag>
mtag='yes'
</tag>
</item>
<item>
<ruleref uri="#MI_3_elem1_rule"/>
<tag>
mtag='no'
</tag>
</item>
</one-of>
</rule>
<rule id="MI_1_elem0_rule" scope="public">
<item>
yes
</item>
</rule>
<rule id="MI_3_elem1_rule" scope="public">
<item>
no
</item>
</rule>
</grammar>

2025-09-07 15:35:05:573920 [INFO] Assign Control Channel <f2981b70f54f4ebe@speechrecog> to Connection 10.80.64.150:1544 <-> 10.80.60.105:54388 [0] -> [1]
2025-09-07 15:35:05:573934 [WARN] Cannot read the first field in start-line
2025-09-07 15:35:05:573937 [WARN] Failed to Parse MRCPv2 Data
2025-09-07 15:35:05:573940 [WARN] Cannot read the first field in start-line
2025-09-07 15:35:05:573955 [INFO] Process DEFINE-GRAMMAR Request <f2981b70f54f4ebe@speechrecog> [100]
2025-09-07 15:35:05:574101 [INFO] Process DEFINE-GRAMMAR Response <f2981b70f54f4ebe@speechrecog> [100]
2025-09-07 15:35:05:574127 [INFO] Send MRCPv2 Data 10.80.64.150:1544 <-> 10.80.60.105:54388 [114 bytes]
MRCP/2.0 114 100 200 COMPLETE
Channel-Identifier: f2981b70f54f4ebe@speechrecog
Completion-Cause: 000 success

2025-09-07 15:35:05:677073 [INFO] Receive MRCPv2 Data 10.80.64.150:1544 <-> 10.80.60.105:54388 [275 bytes]
MRCP/2.0 273 RECOGNIZE 101
Channel-Identifier: f2981b70f54f4ebe@speechrecog
Cancel-If-Queue: false
No-Input-Timeout:10000
Content-Type:text/uri-list
Speech-Language:en-US
Start-Input-Timers:false
Content-Length: 46

session:SimpleRe...@DialogObject.grammar

2025-09-07 15:35:05:677137 [WARN] Cannot read the first field in start-line
2025-09-07 15:35:05:677141 [WARN] Failed to Parse MRCPv2 Data
2025-09-07 15:35:05:677142 [WARN] Cannot read the first field in start-line
2025-09-07 15:35:05:677175 [INFO] Process RECOGNIZE Request <f2981b70f54f4ebe@speechrecog> [101]
2025-09-07 15:35:05:677255 [INFO] Init Speech Detector: frame-duration=10 ms, frame-size=160, max-frame-count=1550, output-frame-count=20, vad-mode=0, noinput-timeout=10000 ms, input-timeout=30000 ms, start-timeout=50 ms, complete-timeout=1000 ms, incomplete-timeout=15000 ms, leading-silence=300 ms, trailing-silence=300 ms, interim-results=1, start-of-input=external <f2981b70f54f4ebe>
2025-09-07 15:35:05:677289 [INFO] Open Waveform File for Writing /opt/unimrcp/var/umsazuresr-f2981b70f54f4ebe-101-8-kHz.wav, sampling-rate [8000]
2025-09-07 15:35:05:677417 [INFO] Select Service Endpoint: method [round-robin] exclude [0] language [en-US] profile []
2025-09-07 15:35:05:677424 [INFO] Set Subscription Profile [cognitive.stt.key] <f2981b70f54f4ebe@azuresr>
2025-09-07 15:35:05:677430 [INFO] Set Recognition Mode [interactive] <f2981b70f54f4ebe@azuresr>
2025-09-07 15:35:05:677476 [INFO] Create Resampler from 8000 Hz to 16000 Hz <f2981b70f54f4ebe@azuresr>
2025-09-07 15:35:05:677508 [INFO] Initiate WS connection <f2981b70f54f4ebe> guid [6d45158fe0214486ae883e90036d3107] uri [http://localhost:5001/speech/recognition/interactive/cognitiveservices/v1?format=detailed&language=en-US]
2025-09-07 15:35:05:677517 [INFO] Open Waveform File for Writing /opt/unimrcp/var/umsazuresr-f2981b70f54f4ebe-101-16-kHz.wav, sampling-rate [16000]
2025-09-07 15:35:05:677583 [INFO] Process RECOGNIZE Response <f2981b70f54f4ebe@speechrecog> [101]
2025-09-07 15:35:05:677587 [INFO] State Transition IDLE -> RECOGNIZING <f2981b70f54f4ebe@speechrecog>
2025-09-07 15:35:05:677604 [INFO] Send MRCPv2 Data 10.80.64.150:1544 <-> 10.80.60.105:54388 [85 bytes]
MRCP/2.0 85 101 200 IN-PROGRESS
Channel-Identifier: f2981b70f54f4ebe@speechrecog

2025-09-07 15:35:05:677630 [INFO] WS connected <f2981b70f54f4ebe>
2025-09-07 15:35:05:677636 [INFO] WS upgrade guid [6d45158fe0214486ae883e90036d3107] <f2981b70f54f4ebe>
2025-09-07 15:35:05:680783 [INFO] WS upgraded <f2981b70f54f4ebe>
HTTP/1.1 101 Switching Protocols
Connection: Upgrade
Date: Sun, 07 Sep 2025 12:35:04 GMT
Server: Kestrel
Upgrade: websocket
Sec-WebSocket-Accept: vSoikejPr3e5BeyEKyn5kfvso1I=

2025-09-07 15:35:05:680833 [INFO] Send WS msg [232 bytes] <f2981b70f54f4ebe>
Path:speech.config
X-Timestamp:2025-09-07T15:35:05.680815
Content-Type:application/json; charset=utf-8

{"context":{"system":{"version":"2.0.12341"}},"device":{"manufacturer":"UniSpeech","model":"umsazuresr","version":"1.32.0"}}
2025-09-07 15:35:08:047707 [INFO] Receive MRCPv2 Data 10.80.64.150:1544 <-> 10.80.60.105:54388 [90 bytes]
MRCP/2.0 88 START-INPUT-TIMERS 102
Channel-Identifier: f2981b70f54f4ebe@speechrecog

2025-09-07 15:35:08:047759 [WARN] Cannot read the first field in start-line
2025-09-07 15:35:08:047781 [INFO] Process START-INPUT-TIMERS Request <f2981b70f54f4ebe@speechrecog> [102]
2025-09-07 15:35:08:047827 [INFO] Start No-Input Timer [10000 ms] <f2981b70f54f4ebe>
2025-09-07 15:35:08:047851 [INFO] Process START-INPUT-TIMERS Response <f2981b70f54f4ebe@speechrecog> [102]
2025-09-07 15:35:08:047873 [INFO] Send MRCPv2 Data 10.80.64.150:1544 <-> 10.80.60.105:54388 [82 bytes]
MRCP/2.0 82 102 200 COMPLETE
Channel-Identifier: f2981b70f54f4ebe@speechrecog

2025-09-07 15:35:18:047357 [INFO] Input Complete [success] size=0 bytes, dur=0 ms <f2981b70f54f4ebe@azuresr>
2025-09-07 15:35:18:047410 [NOTICE] UTTM total file entries [7] timers [7]
2025-09-07 15:35:18:047419 [NOTICE] UTTM total file entries [8] timers [8]
2025-09-07 15:35:18:047590 [NOTICE] RDRM total file entries [4] timers [4]
2025-09-07 15:35:18:047591 [INFO] Process RECOGNITION-COMPLETE Event <f2981b70f54f4ebe@speechrecog> [101]
2025-09-07 15:35:18:047626 [INFO] State Transition RECOGNIZING -> RECOGNIZED <f2981b70f54f4ebe@speechrecog>
2025-09-07 15:35:18:047652 [INFO] Send MRCPv2 Data 10.80.64.150:1544 <-> 10.80.60.105:54388 [140 bytes]
MRCP/2.0 140 RECOGNITION-COMPLETE 101 COMPLETE
Channel-Identifier: f2981b70f54f4ebe@speechrecog
Completion-Cause: 002 no-input-timeout

2025-09-07 15:35:18:063759 [INFO] Receive MRCPv2 Data 10.80.64.150:1544 <-> 10.80.60.105:54388 [763 bytes]
MRCP/2.0 761 DEFINE-GRAMMAR 103
Channel-Identifier: f2981b70f54f4ebe@speechrecog
Content-Type:application/srgs+xml
Speech-Language:en-US
Content-Id:SimpleRe...@DialogObject.grammar
Content-Length: 544

<?xml version="1.0" encoding="UTF-8"?>
<grammar root="main" tag-format="swi-semantics/1.0" version="1.0" xml:lang="en-US" xmlns="http://www.w3.org/2001/06/grammar">
<rule id="main" scope="public">
<one-of>
<item>
<ruleref uri="#MI_1_elem0_rule"/>
<tag>
mtag='yes'
</tag>
</item>
<item>
<ruleref uri="#MI_3_elem1_rule"/>
<tag>
mtag='no'
</tag>
</item>
</one-of>
</rule>
<rule id="MI_1_elem0_rule" scope="public">
<item>
yes
</item>
</rule>
<rule id="MI_3_elem1_rule" scope="public">
<item>
no
</item>
</rule>
</grammar>

2025-09-07 15:35:18:063813 [WARN] Cannot read the first field in start-line
2025-09-07 15:35:18:063816 [WARN] Failed to Parse MRCPv2 Data
2025-09-07 15:35:18:063818 [WARN] Cannot read the first field in start-line
2025-09-07 15:35:18:063850 [INFO] Process DEFINE-GRAMMAR Request <f2981b70f54f4ebe@speechrecog> [103]
2025-09-07 15:35:18:063868 [INFO] State Transition RECOGNIZED -> IDLE <f2981b70f54f4ebe@speechrecog>
2025-09-07 15:35:18:064030 [INFO] Process DEFINE-GRAMMAR Response <f2981b70f54f4ebe@speechrecog> [103]
2025-09-07 15:35:18:064055 [INFO] Send MRCPv2 Data 10.80.64.150:1544 <-> 10.80.60.105:54388 [114 bytes]
MRCP/2.0 114 103 200 COMPLETE
Channel-Identifier: f2981b70f54f4ebe@speechrecog
Completion-Cause: 000 success

2025-09-07 15:35:18:067620 [INFO] Receive MRCPv2 Data 10.80.64.150:1544 <-> 10.80.60.105:54388 [275 bytes]
MRCP/2.0 273 RECOGNIZE 104
Channel-Identifier: f2981b70f54f4ebe@speechrecog
Cancel-If-Queue: false
No-Input-Timeout:10000
Content-Type:text/uri-list
Speech-Language:en-US
Start-Input-Timers:false
Content-Length: 46

session:SimpleRe...@DialogObject.grammar

2025-09-07 15:35:18:067672 [WARN] Cannot read the first field in start-line
2025-09-07 15:35:18:067675 [WARN] Failed to Parse MRCPv2 Data
2025-09-07 15:35:18:067676 [WARN] Cannot read the first field in start-line
2025-09-07 15:35:18:067691 [INFO] Process RECOGNIZE Request <f2981b70f54f4ebe@speechrecog> [104]
2025-09-07 15:35:18:067743 [INFO] Init Speech Detector: frame-duration=10 ms, frame-size=160, max-frame-count=1550, output-frame-count=20, vad-mode=0, noinput-timeout=10000 ms, input-timeout=30000 ms, start-timeout=50 ms, complete-timeout=1000 ms, incomplete-timeout=15000 ms, leading-silence=300 ms, trailing-silence=300 ms, interim-results=1, start-of-input=external <f2981b70f54f4ebe>
2025-09-07 15:35:18:067796 [INFO] Open Waveform File for Writing /opt/unimrcp/var/umsazuresr-f2981b70f54f4ebe-104-8-kHz.wav, sampling-rate [8000]
2025-09-07 15:35:18:067855 [INFO] Set Recognition Mode [interactive] <f2981b70f54f4ebe@azuresr>
2025-09-07 15:35:18:067891 [INFO] Reset Resampler <f2981b70f54f4ebe@azuresr>
2025-09-07 15:35:18:067896 [INFO] Open Waveform File for Writing /opt/unimrcp/var/umsazuresr-f2981b70f54f4ebe-104-16-kHz.wav, sampling-rate [16000]
2025-09-07 15:35:18:067932 [INFO] Process RECOGNIZE Response <f2981b70f54f4ebe@speechrecog> [104]
2025-09-07 15:35:18:067936 [INFO] State Transition IDLE -> RECOGNIZING <f2981b70f54f4ebe@speechrecog>
2025-09-07 15:35:18:067952 [INFO] Send MRCPv2 Data 10.80.64.150:1544 <-> 10.80.60.105:54388 [85 bytes]
MRCP/2.0 85 104 200 IN-PROGRESS
Channel-Identifier: f2981b70f54f4ebe@speechrecog

2025-09-07 15:35:20:276951 [INFO] Receive MRCPv2 Data 10.80.64.150:1544 <-> 10.80.60.105:54388 [76 bytes]
MRCP/2.0 74 STOP 105
Channel-Identifier: f2981b70f54f4ebe@speechrecog

2025-09-07 15:35:20:277003 [WARN] Cannot read the first field in start-line
2025-09-07 15:35:20:277048 [INFO] Process STOP Request <f2981b70f54f4ebe@speechrecog> [105]
2025-09-07 15:35:20:277213 [INFO] Close HTTP connection <f2981b70f54f4ebe>
2025-09-07 15:35:20:277303 [INFO] Input Complete [stopped] size=0 bytes, dur=0 ms <f2981b70f54f4ebe@azuresr>
2025-09-07 15:35:20:277348 [NOTICE] UTTM total file entries [9] timers [9]
2025-09-07 15:35:20:277360 [NOTICE] UTTM total file entries [10] timers [10]
2025-09-07 15:35:20:277517 [INFO] Process STOP Response <f2981b70f54f4ebe@speechrecog> [105]
2025-09-07 15:35:20:277526 [INFO] State Transition RECOGNIZING -> IDLE <f2981b70f54f4ebe@speechrecog>
2025-09-07 15:35:20:277530 [NOTICE] RDRM total file entries [5] timers [5]
2025-09-07 15:35:20:277548 [INFO] Send MRCPv2 Data 10.80.64.150:1544 <-> 10.80.60.105:54388 [112 bytes]
MRCP/2.0 112 105 200 COMPLETE
Channel-Identifier: f2981b70f54f4ebe@speechrecog
Active-Request-Id-List: 104

2025-09-07 15:35:21:307585 [INFO] Receive MRCPv2 Data 10.80.64.150:1544 <-> 10.80.60.105:54388 [76 bytes]
MRCP/2.0 74 STOP 106
Channel-Identifier: f2981b70f54f4ebe@speechrecog

2025-09-07 15:35:21:307656 [WARN] Cannot read the first field in start-line
2025-09-07 15:35:21:307732 [INFO] Process STOP Request <f2981b70f54f4ebe@speechrecog> [106]
2025-09-07 15:35:21:307774 [INFO] Send MRCPv2 Data 10.80.64.150:1544 <-> 10.80.60.105:54388 [82 bytes]
MRCP/2.0 82 106 200 COMPLETE
Channel-Identifier: f2981b70f54f4ebe@speechrecog

2025-09-07 15:35:21:328388 [INFO] TCP/MRCPv2 Peer Disconnected 10.80.64.150:1544 <-> 10.80.60.105:54388
2025-09-07 15:35:21:336435 [INFO] Receive SIP Event [nua_i_bye] Status 200 Session Terminated [SIP-Agent-1]
2025-09-07 15:35:21:336452 [INFO] Receive SIP Event [nua_i_state] Status 200 Session Terminated [SIP-Agent-1]
2025-09-07 15:35:21:336458 [NOTICE] SIP Call State 0x7f6c84002808 [terminated]
2025-09-07 15:35:21:336473 [INFO] Receive SIP Event [nua_i_terminated] Status 200 Session Terminated [SIP-Agent-1]
2025-09-07 15:35:21:336833 [INFO] Deactivate Session 0x7f6c84002808 <f2981b70f54f4ebe>
2025-09-07 15:35:21:336848 [INFO] Terminate Session 0x7f6c84002808 <f2981b70f54f4ebe>
2025-09-07 15:35:21:336887 [INFO] Remove Control Channel <f2981b70f54f4ebe@speechrecog> [0]
2025-09-07 15:35:21:336907 [INFO] Close <f2981b70f54f4ebe@azuresr>
2025-09-07 15:35:21:336958 [INFO] Close HTTP connection <f2981b70f54f4ebe>
2025-09-07 15:35:21:337013 [NOTICE] AZURESR Usage: 0/1/2
2025-09-07 15:35:21:337050 [NOTICE] Usage [umsazuresr] min [0] cur [0] max [1]
2025-09-07 15:35:21:337274 [INFO] Close RTP Receiver 10.80.64.150:4002 <- 10.80.60.105:25448 [r:157 l:0 j:0 p:50 d:0 i:157]
2025-09-07 15:35:21:338714 [INFO] Remove RTP Session 10.80.64.150:4002
2025-09-07 15:35:21:338719 [INFO] Generate RTCP RR [ssrc:2097 last_seq:2258 j:0 lost:0 frac:0]
2025-09-07 15:35:21:338722 [INFO] Send Compound RTCP Packet [BYE] [80 bytes] 10.80.64.150:4003 -> 10.80.60.105:25449
2025-09-07 15:35:21:338794 [NOTICE] Destroy TCP/MRCPv2 Connection 10.80.64.150:1544 <-> 10.80.60.105:54388
2025-09-07 15:35:21:338815 [NOTICE] Remove Session <f2981b70f54f4ebe>
2025-09-07 15:35:21:338818 [INFO] Session Terminated 0x7f6c84002808 <f2981b70f54f4ebe>

2025-09-07 15:35:21:338840 [NOTICE] Destroy Session <f2981b70f54f4ebe>

Daniel Friedman

unread,

Sep 16, 2025, 3:41:05 AMSep 16

to UniMRCP

Hello,

i would like to hear from your support team, as it is holding back our project.

please let us know how can we solve this problem of the RTP not flowing from the uniMRCP to the Azure SR plugin.

we tested it locally with the ums client and it is working with no problem.

Thanks,

Dani

Sunny Zhang

unread,

Oct 2, 2025, 4:39:21 AMOct 2

to UniMRCP

Hi Daniel,

I have a similar environment to yours. Would you mind sharing your UCCX and UniMRCP config with me? I am now having trouble connecting to UCCX using demo TTS/ASR in UniMRCP.

Thanks.

SunnyZhang2

Reply all

Reply to author

Forward

0 new messages