​
Overview

Rime AI’s text-to-speech capabilities are available through two service implementations:

  • RimeTTSService: WebSocket-based implementation with word-level timing and interruption support
  • RimeHttpTTSService: HTTP-based implementation for simpler use cases

You can obtain a Rime API key by signing up at Rime.

​
RimeTTSService (WebSocket Service)

Uses Rime’s WebSocket JSON API for real-time speech synthesis with word-level timing information.

​
Constructor Parameters

​
api_key
str
required

Rime API key

​
voice_id
str
required

Rime voice identifier

​
url
str
default:"wss://users-ws.rime.ai/ws2"

Rime WebSocket API endpoint

​
model
str
default:"mistv2"

Model ID to use for synthesis

​
sample_rate
int
default:"None"

Output audio sample rate in Hz

​
params
InputParams
default:"InputParams()"

Speech generation parameters

​
Features

  • Word-level timing information
  • Support for interruptions
  • Context tracking across multiple messages
  • Real-time audio streaming
  • Proper sentence aggregation

​
RimeHttpTTSService (HTTP Service)

​
Constructor Parameters

​
api_key
str
required

Rime API key

​
voice_id
str
default:"eva"

Rime voice identifier. See Rime’s documentation for supported voices.

​
model
str
default:"mist"

Choose mist for hyper-realistic conversational voices or v1 for Rime’s first-gen model.

​
sample_rate
int
default:"None"

Output audio sample rate in Hz

​
params
InputParams
default:"InputParams()"

Speech generation parameters

​
Output Frames

Both services generate the following frames:

​
Control Frames

​
TTSStartedFrame
Frame

Signals start of speech synthesis

​
TTSStoppedFrame
Frame

Signals completion of speech synthesis

​
Audio Frames

​
TTSAudioRawFrame
Frame

Contains generated audio data: - PCM audio format - Specified sample rate - Single channel (mono)

​
Text Frames (WebSocket only)

​
TTSTextFrame
Frame

Contains word-level text with timing information

​
Error Frames

​
ErrorFrame
Frame

Contains Rime TTS error information

​
Usage Example

# WebSocket Service
from pipecat.services.rime import RimeTTSService

ws_tts = RimeTTSService(
    api_key="your-rime-api-key",
    voice_id="cove",
    model="mistv2",
    params=RimeTTSService.InputParams(
        language=Language.EN,
        speed_alpha=1.0
    )
)

# HTTP Service
from pipecat.services.rime import RimeHttpTTSService

http_tts = RimeHttpTTSService(
    api_key="your-rime-api-key",
    voice_id="eva",
    model="mist",
    params=RimeHttpTTSService.InputParams(
        speed_alpha=1.2,
        reduce_latency=True
    )
)

# Use in pipeline
pipeline = Pipeline([
    ...,
    llm,
    ws_tts,  # or http_tts
    transport.output(),
])

​
Frame Flow

TextFrame

RimeTTSService

TTSStartedFrame

TTSAudioRawFrame

TTSTextFrame

TTSStoppedFrame

ErrorFrame

​
Metrics Support

Both services collect processing metrics:

  • Time to First Byte (TTFB)
  • Character usage statistics

​
Service Comparison

FeatureWebSocketHTTP
Word timing✓-
Interruption support✓-
Bracket-based pauses-✓
Phoneme control-✓
Inline speed control-✓
Streaming audio✓✓
PlayHTXTTS