Skip to content

Commit bb1f3e9

Browse files
Google APIscopybara-github
Google APIs
authored and
copybara-github
committed
docs: clarified meaning of the legacy editions
docs: clarified semantic of the streaming APIs PiperOrigin-RevId: 404815104
1 parent f034e5f commit bb1f3e9

3 files changed

Lines changed: 68 additions & 62 deletions

File tree

‎google/cloud/dialogflow/v2/agent.proto‎

Lines changed: 5 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -293,14 +293,15 @@ message Agent {
293293
// Not specified. This value should never be used.
294294
TIER_UNSPECIFIED = 0;
295295

296-
// Standard tier.
296+
// Trial Edition, previously known as Standard Edition.
297297
TIER_STANDARD = 1;
298298

299-
// Enterprise tier (Essentials).
299+
// Essentials Edition, previously known as Enterprise Essential Edition.
300300
TIER_ENTERPRISE = 2;
301301

302-
// Enterprise tier (Plus).
303-
TIER_ENTERPRISE_PLUS = 3;
302+
// Essentials Edition (same as TIER_ENTERPRISE), previously known as
303+
// Enterprise Plus Edition.
304+
TIER_ENTERPRISE_PLUS = 3 [deprecated = true];
304305
}
305306

306307
// Required. The project of this agent.

‎google/cloud/dialogflow/v2/audio_config.proto‎

Lines changed: 30 additions & 30 deletions
Original file line numberDiff line numberDiff line change
@@ -30,36 +30,6 @@ option java_outer_classname = "AudioConfigProto";
3030
option java_package = "com.google.cloud.dialogflow.v2";
3131
option objc_class_prefix = "DF";
3232

33-
// Hints for the speech recognizer to help with recognition in a specific
34-
// conversation state.
35-
message SpeechContext {
36-
// Optional. A list of strings containing words and phrases that the speech
37-
// recognizer should recognize with higher likelihood.
38-
//
39-
// This list can be used to:
40-
//
41-
// * improve accuracy for words and phrases you expect the user to say,
42-
// e.g. typical commands for your Dialogflow agent
43-
// * add additional words to the speech recognizer vocabulary
44-
// * ...
45-
//
46-
// See the [Cloud Speech
47-
// documentation](https://cloud.google.com/speech-to-text/quotas) for usage
48-
// limits.
49-
repeated string phrases = 1;
50-
51-
// Optional. Boost for this context compared to other contexts:
52-
//
53-
// * If the boost is positive, Dialogflow will increase the probability that
54-
// the phrases in this context are recognized over similar sounding phrases.
55-
// * If the boost is unspecified or non-positive, Dialogflow will not apply
56-
// any boost.
57-
//
58-
// Dialogflow recommends that you use boosts in the range (0, 20] and that you
59-
// find a value that fits your use case with binary search.
60-
float boost = 2;
61-
}
62-
6333
// Audio encoding of the audio content sent in the conversational query request.
6434
// Refer to the
6535
// [Cloud Speech API
@@ -109,6 +79,36 @@ enum AudioEncoding {
10979
AUDIO_ENCODING_SPEEX_WITH_HEADER_BYTE = 7;
11080
}
11181

82+
// Hints for the speech recognizer to help with recognition in a specific
83+
// conversation state.
84+
message SpeechContext {
85+
// Optional. A list of strings containing words and phrases that the speech
86+
// recognizer should recognize with higher likelihood.
87+
//
88+
// This list can be used to:
89+
//
90+
// * improve accuracy for words and phrases you expect the user to say,
91+
// e.g. typical commands for your Dialogflow agent
92+
// * add additional words to the speech recognizer vocabulary
93+
// * ...
94+
//
95+
// See the [Cloud Speech
96+
// documentation](https://cloud.google.com/speech-to-text/quotas) for usage
97+
// limits.
98+
repeated string phrases = 1;
99+
100+
// Optional. Boost for this context compared to other contexts:
101+
//
102+
// * If the boost is positive, Dialogflow will increase the probability that
103+
// the phrases in this context are recognized over similar sounding phrases.
104+
// * If the boost is unspecified or non-positive, Dialogflow will not apply
105+
// any boost.
106+
//
107+
// Dialogflow recommends that you use boosts in the range (0, 20] and that you
108+
// find a value that fits your use case with binary search.
109+
float boost = 2;
110+
}
111+
112112
// Information for a word recognized by the speech recognizer.
113113
message SpeechWordInfo {
114114
// The word this info is for.

‎google/cloud/dialogflow/v2/session.proto‎

Lines changed: 33 additions & 28 deletions
Original file line numberDiff line numberDiff line change
@@ -479,10 +479,11 @@ message StreamingDetectIntentRequest {
479479
//
480480
// Multiple response messages can be returned in order:
481481
//
482-
// 1. If the input was set to streaming audio, the first one or more messages
483-
// contain `recognition_result`. Each `recognition_result` represents a more
484-
// complete transcript of what the user said. The last `recognition_result`
485-
// has `is_final` set to `true`.
482+
// 1. If the `StreamingDetectIntentRequest.input_audio` field was
483+
// set, the `recognition_result` field is populated for one
484+
// or more messages.
485+
// See the [StreamingRecognitionResult][google.cloud.dialogflow.v2.StreamingRecognitionResult] message for details
486+
// about the result message sequence.
486487
//
487488
// 2. The next message contains `response_id`, `query_result`
488489
// and optionally `webhook_status` if a WebHook was called.
@@ -520,35 +521,39 @@ message StreamingDetectIntentResponse {
520521
// that is currently being processed or an indication that this is the end
521522
// of the single requested utterance.
522523
//
523-
// Example:
524+
// While end-user audio is being processed, Dialogflow sends a series of
525+
// results. Each result may contain a `transcript` value. A transcript
526+
// represents a portion of the utterance. While the recognizer is processing
527+
// audio, transcript values may be interim values or finalized values.
528+
// Once a transcript is finalized, the `is_final` value is set to true and
529+
// processing continues for the next transcript.
524530
//
525-
// 1. transcript: "tube"
531+
// If `StreamingDetectIntentRequest.query_input.audio_config.single_utterance`
532+
// was true, and the recognizer has completed processing audio,
533+
// the `message_type` value is set to `END_OF_SINGLE_UTTERANCE and the
534+
// following (last) result contains the last finalized transcript.
526535
//
527-
// 2. transcript: "to be a"
536+
// The complete end-user utterance is determined by concatenating the
537+
// finalized transcript values received for the series of results.
528538
//
529-
// 3. transcript: "to be"
539+
// In the following example, single utterance is enabled. In the case where
540+
// single utterance is not enabled, result 7 would not occur.
530541
//
531-
// 4. transcript: "to be or not to be"
532-
// is_final: true
542+
// ```
543+
// Num | transcript | message_type | is_final
544+
// --- | ----------------------- | ----------------------- | --------
545+
// 1 | "tube" | TRANSCRIPT | false
546+
// 2 | "to be a" | TRANSCRIPT | false
547+
// 3 | "to be" | TRANSCRIPT | false
548+
// 4 | "to be or not to be" | TRANSCRIPT | true
549+
// 5 | "that's" | TRANSCRIPT | false
550+
// 6 | "that is | TRANSCRIPT | false
551+
// 7 | unset | END_OF_SINGLE_UTTERANCE | unset
552+
// 8 | " that is the question" | TRANSCRIPT | true
553+
// ```
533554
//
534-
// 5. transcript: " that's"
535-
//
536-
// 6. transcript: " that is"
537-
//
538-
// 7. message_type: `END_OF_SINGLE_UTTERANCE`
539-
//
540-
// 8. transcript: " that is the question"
541-
// is_final: true
542-
//
543-
// Only two of the responses contain final results (#4 and #8 indicated by
544-
// `is_final: true`). Concatenating these generates the full transcript: "to be
545-
// or not to be that is the question".
546-
//
547-
// In each response we populate:
548-
//
549-
// * for `TRANSCRIPT`: `transcript` and possibly `is_final`.
550-
//
551-
// * for `END_OF_SINGLE_UTTERANCE`: only `message_type`.
555+
// Concatenating the finalized transcripts with `is_final` set to true,
556+
// the complete utterance becomes "to be or not to be that is the question".
552557
message StreamingRecognitionResult {
553558
// Type of the response message.
554559
enum MessageType {

0 commit comments

Comments
 (0)